UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Spatial and temporal patterns of sediment mobility and storage in a small mountain stream Klinghoffer, Ilana 2015

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2015_may_klinghoffer_ilana.pdf [ 9.91MB ]
Metadata
JSON: 24-1.0167165.json
JSON-LD: 24-1.0167165-ld.json
RDF/XML (Pretty): 24-1.0167165-rdf.xml
RDF/JSON: 24-1.0167165-rdf.json
Turtle: 24-1.0167165-turtle.txt
N-Triples: 24-1.0167165-rdf-ntriples.txt
Original Record: 24-1.0167165-source.json
Full Text
24-1.0167165-fulltext.txt
Citation
24-1.0167165.ris

Full Text

Spatial and Temporal Patterns of Sediment Mobility andStorage in a Small Mountain StreambyIlana KlinghofferBSc in Environmental Sciences, University of Guelph, 2010A THESIS SUBMITTED IN PARTIAL FULFILLMENTOF THE REQUIREMENTS FOR THE DEGREE OFMaster of ScienceinTHE FACULTY OF GRADUATE AND POSTDOCTORALSTUDIES(Geography)The University of British Columbia(Vancouver)March 2015c© Ilana Klinghoffer, 2015AbstractThe study was conducted in East Creek, a headwater gravel-bed channel in theFraser Valley foothills of the Coastal Mountains of British Columbia. Sedimenttransport was measured at three spatial scales using two measurement techniquesin a study reach containing three unique morphological reaches: rapids, riffle pool,and step pool. At the largest spatial scale, the channel scale, channel stability wasassessed between 2003 and 2009 using longitudinal profiles of channel elevationobtained from digital elevation mapping. The longitudinal profiles suggest thatEast Creek was in a relatively stable state over the six year analysis period, withthe majority of erosion and deposition limited to localized fluctuations that var-ied in magnitude and direction. At the intermediate spatial scale, the reach scale,sediment transport estimates obtained from pit trap and digital elevation mappingdata were used to create a sediment budget for the rapids reach and riffle pool sub-reaches of the channel. Using both measurement techniques, erosion and deposi-tion fluctuated and could not be linked to flow regime or sediment supply alone.It is hypothesized that in-stream sediment supply and bed conditioning are impor-tant controls on sediment storage, and were used to explain observed fluctuationsin erosion and deposition. The magnitude and direction of reach scale sedimentstorage fluctuations were not consistent across the two measurement techniques;however, elevation mapping estimates were nearly always higher than pit trap esti-mates. This is likely a result of overpassing of fine material and pit trap inefficiency.At the smallest spatial scale, the unit scale, spatial patterns of sediment transportwere assessed across riffles and pools using digital elevation and morphologicalmapping data. There was increased sediment mobility in pools compared to riffles,which is likely a result of pools containing finer more loosely interacting particlesiicompared to those in riffles. The high resolution unit scale sediment storage datademonstrated conservation of mass and a tight coupling of erosion and depositionin East Creek.iiiPrefaceThis dissertation is based on data collected in East Creek between 2003 and 2011under the direction of Marwan Hassan. Given the collaborative nature of the projectand the high resolution and long term qualities of the data set, there were a numberof people who made significant contributions to the data collection, compilation,and analysis.Marwan Hassan led and coordinated the East Creek database creation and man-agement. Marwan managed field personnel in the collection, compilation, andanalysis of data in East Creek for the entire duration of the study period includingthe initial set up of equipment at East Creek.Joshua Caulkins performed a large portion of the fieldwork that generated thedatabase analyzed in this thesis, including total station surveying, sediment collec-tion from traps, sediment weighing, aerial photography, and discharge measure-ments. Joshua Caulkins also performed analyses that I built upon in this thesis,including processing and analyzing survey data, rectifying aerial photographs, andcreating the initial morphology map that I later modified and used to demarcatepools and riffles. Joshua Caulkins also provided details on field equipment specifi-cations and extensive information on East Creek channel characteristics.Dave Reid was heavily involved in field data collection including total stationsurveying, sediment collection from traps, sediment weighing, and aerial photog-raphy. Dave was also involved in survey data processing.Other field personnel involved in collecting data used in this thesis include An-dre Zimmermann, Tony Lagemaat, Sam Robinson, Michael More, and Tim Reid.There were additional research assistants who had varying degrees of involvementin field data collection over the eight year study period.ivI used a MATLAB program created by Joseph Wheaton to generate differencemaps of elevation change in East Creek available for public download from JosephWheaton’s website at: http://gcd.joewheaton.org/.I used a MATLAB program created by Shawn Chartrand to generate erosionand deposition histograms for pools and riffles in East Creek.I participated in field data collection in East Creek between 2008 and 2011,including total station surveying, sediment collection from traps, sediment weigh-ing, aerial photography, and discharge measurements. I contributed to the process-ing of data collected between 2006 and 2008, and processed the majority of datacollected between 2008 and 2011, including survey data compilation, aerial pho-tograph rectification, and pit trap data compilation. Unless otherwise noted, theanalyses described in this document were performed by me.vTable of ContentsAbstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iiPreface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ivTable of Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viList of Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ixList of Figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiiAcknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xvi1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.1 Background Information . . . . . . . . . . . . . . . . . . . . . . 11.1.1 Channel Stability and Adjustment . . . . . . . . . . . . . 21.1.2 Channel Morphology . . . . . . . . . . . . . . . . . . . . 31.1.3 Sediment Transport and Storage . . . . . . . . . . . . . . 51.1.4 Bed load Storage Quantification . . . . . . . . . . . . . . 71.2 Research Gap and Study Objectives . . . . . . . . . . . . . . . . 82 Study Site . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92.1 Top of Rapids . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112.2 Rapids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112.3 Riffle Pool . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122.4 Step Pool . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14vi3 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153.1 Data Collection . . . . . . . . . . . . . . . . . . . . . . . . . . . 153.1.1 Pit Traps . . . . . . . . . . . . . . . . . . . . . . . . . . 153.1.2 Ground Surveys . . . . . . . . . . . . . . . . . . . . . . . 163.1.3 Aerial Photography . . . . . . . . . . . . . . . . . . . . . 173.2 Data Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173.2.1 Storage Estimates using Pit Traps . . . . . . . . . . . . . 183.2.2 Storage Estimates using Ground Surveys . . . . . . . . . 183.2.3 Morphology Mapping using Aerial Photography . . . . . 214 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 224.1 Flow Regime . . . . . . . . . . . . . . . . . . . . . . . . . . . . 224.2 Channel Scale: Trends in Sediment Storage . . . . . . . . . . . . 234.3 Reach Scale: Sediment Flux and Storage . . . . . . . . . . . . . . 294.3.1 Bed load Transport from Pit Traps . . . . . . . . . . . . . 294.3.2 Bed Material Transfer from Morphological Method . . . . 304.4 Unit-Scale: Distribution of Sediment Storage . . . . . . . . . . . 384.4.1 Bed Elevation Change Histograms . . . . . . . . . . . . . 384.4.2 Erosion and Deposition Histograms . . . . . . . . . . . . 424.4.3 Bed Storage, Erosion, and Deposition . . . . . . . . . . . 504.4.4 Peak Discharge, Erosion, and Deposition . . . . . . . . . 525 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 565.1 Flow Regime . . . . . . . . . . . . . . . . . . . . . . . . . . . . 565.2 Channel Scale: Long Profile . . . . . . . . . . . . . . . . . . . . 575.3 Reach Scale: Closing the Sediment Budget . . . . . . . . . . . . 585.3.1 Bed Load Flux and Storage . . . . . . . . . . . . . . . . . 585.3.2 Bed Material Erosion and Storage . . . . . . . . . . . . . 605.3.3 Comparison of Methods . . . . . . . . . . . . . . . . . . 635.4 Unit Scale: Sediment Storage and Bedforms . . . . . . . . . . . . 675.4.1 Bed Elevation Change in Riffles and Pools . . . . . . . . 675.4.2 Erosion and Deposition in Riffles and Pools . . . . . . . . 715.4.3 Bed Storage, Erosion, and Deposition . . . . . . . . . . . 75vii5.4.4 Peak Discharge, Erosion, and Deposition . . . . . . . . . 756 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78A Supporting Results: Bed Elevation Change . . . . . . . . . . . . . . 83A.1 Histograms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84A.2 Summary Statistics . . . . . . . . . . . . . . . . . . . . . . . . . 88A.3 Skew . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91B Supporting Results: Erosion and Deposition . . . . . . . . . . . . . 92B.1 Histograms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93B.2 Summary Statistics . . . . . . . . . . . . . . . . . . . . . . . . . 97viiiList of TablesTable 3.1 Pit trap volumes and estimated sediment trapping capacities . . 16Table 3.2 Values of constants used to calculate bed material erosion masses 20Table 4.1 Morphology at areas of interest based on longitudinal profilesof East Creek upper reaches between 2003 and 2009 . . . . . . 27Table 4.2 Bed load flux estimates using pit traps for WY09-11 . . . . . . 30Table 4.3 Bed load storage based on pit traps for WY09-11 . . . . . . . . 31Table 4.4 Bed material export estimates using morphological methods forWY04-11 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32Table 4.5 Summary statistics for bed elevation change in riffles in rifflepool 3 sub-reach . . . . . . . . . . . . . . . . . . . . . . . . . 40Table 4.6 Summary statistics for bed elevation change in pools in rifflepool 3 sub-reach . . . . . . . . . . . . . . . . . . . . . . . . . 42Table 4.7 Summary statistics for erosion in riffles in riffle pool 2 sub-reach 46Table 4.8 Summary statistics for deposition in riffles in riffle pool 2 sub-reach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46Table 4.9 Summary statistics for erosion in pools in riffle pool 2 sub-reach 50Table 4.10 Summary statistics for deposition in pools in riffle pool 2 sub-reach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51Table 4.11 Correlation between peak discharge and mean bed deposition . 54Table 4.12 Correlation between peak discharge and mean bed erosion . . . 55Table 5.1 Sediment budget from traps and survey data for East Creek forW09-11 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64ixTable 5.2 Correlation between bed elevation change in riffles and bed el-evation change in pools in upper reaches of East Creek . . . . . 68Table 5.3 Correlation between erosion/deposition in riffles and erosion/de-position in pools in upper reaches of East Creek . . . . . . . . 71Table A.1 Summary statistics for bed elevation change in riffles in rapidsreach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88Table A.2 Summary statistics for bed elevation change in riffles in rifflepool 1 sub-reach . . . . . . . . . . . . . . . . . . . . . . . . . 88Table A.3 Summary statistics for bed elevation change in riffles in rifflepool 2 sub-reach . . . . . . . . . . . . . . . . . . . . . . . . . 89Table A.4 Summary statistics for bed elevation change in pools in rapidsreach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89Table A.5 Summary statistics for bed elevation change in pools in rifflepool 1 sub-reach . . . . . . . . . . . . . . . . . . . . . . . . . 90Table A.6 Summary statistics for bed elevation change in pools in rifflepool 2 sub-reach . . . . . . . . . . . . . . . . . . . . . . . . . 90Table B.1 Summary statistics for erosion in riffles in the rapids reach . . . 97Table B.2 Summary statistics for erosion in riffles in riffle pool 1 sub-reach 97Table B.3 Summary statistics for erosion in riffles in riffle pool 3 sub-reach 98Table B.4 Summary statistics for deposition in riffles in the rapids reach . 98Table B.5 Summary statistics for deposition in riffles in riffle pool 1 sub-reach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99Table B.6 Summary statistics for deposition in riffles in riffle pool 3 sub-reach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99Table B.7 Summary statistics for erosion in pools in rapids reach . . . . . 100Table B.8 Summary statistics for erosion in pools in riffle pool 1 sub-reach 100Table B.9 Summary statistics for erosion in pools in riffle pool 3 sub-reach 101Table B.10 Summary statistics for deposition in pools in rapids reach . . . 101Table B.11 Summary statistics for deposition in pools in riffle pool 1 sub-reach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102xTable B.12 Summary statistics for deposition in pools in riffle pool 3 sub-reach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102xiList of FiguresFigure 2.1 Location of UBC Malcolm Knapp Research Forest in FraserValley Foothills, British Columbia. The research forest is out-lined in purple. (adapted from Google Maps) . . . . . . . . . 10Figure 2.2 Location of rapids, riffle pool 1, riffle pool 2, riffle pool 3, andstep pool in East Creek, British Columbia . . . . . . . . . . . 11Figure 2.3 Photograph of a pit trap outlined in red, located at the bound-ary between the rapids and riffle pool 1 in East Creek, BritishColumbia . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13Figure 2.4 Aerial photograph of riffle pool 1 sub-reach in East Creek takenusing pole photography method . . . . . . . . . . . . . . . . 14Figure 4.1 Annual peak discharge in East Creek for WY04-11 . . . . . . 23Figure 4.2 Longitudinal profile at the channel scale for the rapids and rif-fle pool reaches of East Creek for 2003 to 2009 . . . . . . . . 24Figure 4.3 Longitudinal profiles at the reach scale for the rapids and rifflepool reaches of East Creek for 2003 to 2009 . . . . . . . . . . 25Figure 4.4 Change in channel gradient over time across the upper reachesof East Creek from 2003 to 2009 . . . . . . . . . . . . . . . . 28Figure 4.5 Volumetric change in sediment storage in the rapids reach andriffle pool sub-reaches of East Creek for WY04-11 . . . . . . 34Figure 4.6 Volumetric change in sediment storage moving downstreamalong the bed of the East Creek study reach for WY04-07 . . 36Figure 4.7 Volumetric change in sediment storage moving downstreamalong the bed of the East Creek study reach for WY08-11 . . 37xiiFigure 4.8 Examples of types of bed elevation change histograms in rifflesin East Creek . . . . . . . . . . . . . . . . . . . . . . . . . . 38Figure 4.9 Examples of types of bed elevation change histograms in poolsin East Creek . . . . . . . . . . . . . . . . . . . . . . . . . . 41Figure 4.10 Uni-modal erosion and deposition histograms in riffle pool 3for riffles in WY09 . . . . . . . . . . . . . . . . . . . . . . . 43Figure 4.11 A bi-modal erosion histogram concurrent with a uni-modal de-position histogram in riffle pool 1 for riffles in WY07 . . . . . 44Figure 4.12 Multimodal erosion and deposition histograms in the rapids forriffles in WY10 . . . . . . . . . . . . . . . . . . . . . . . . . 44Figure 4.13 Uni-modal erosion and deposition histograms in riffle pool 3for pools in WY09 . . . . . . . . . . . . . . . . . . . . . . . 48Figure 4.14 A multi-modal erosion histogram concurrent with a bi-modaldeposition histogram in the rapids for pools in WY06 . . . . . 48Figure 4.15 Multi-modal erosion and uni-modal deposition histograms inriffle pool 2 for pools in WY05 . . . . . . . . . . . . . . . . . 49Figure 4.16 Sediment storage vs. erosion and deposition in riffles in com-bined upper reaches of East Creek . . . . . . . . . . . . . . . 51Figure 4.17 Sediment storage vs. erosion and deposition in pools in com-bined upper reaches of East Creek . . . . . . . . . . . . . . . 52Figure 4.18 Peak discharge vs. mean bed erosion and mean bed depositionin riffles of riffle pool 3 sub-reach in East Creek . . . . . . . . 53Figure 4.19 Peak discharge vs. mean bed erosion and mean bed depositionin pools of rapids reach in East Creek . . . . . . . . . . . . . 54Figure 5.1 Upper reaches of East Creek with location of top of rapids(TOR) sub-reach highlighted in red circle . . . . . . . . . . . 58Figure 5.2 Plunge pool at base of culvert in top of rapids (TOR) sub-reachof East Creek . . . . . . . . . . . . . . . . . . . . . . . . . . 59Figure 5.3 Exposure of till on bed of rapids reach of East Creek in WY07following high magnitude scouring . . . . . . . . . . . . . . 61Figure 5.4 Sediment storage estimates from surveys compared to sedi-ment storage estimates from traps . . . . . . . . . . . . . . . 65xiiiFigure 5.5 Difference in sediment storage estimates between two distinctmethods (surveys and traps) scaled to sediment storage esti-mates from surveys . . . . . . . . . . . . . . . . . . . . . . . 67Figure 5.6 Variance in bed elevation over time in riffles (purple line) andpools (blue line) in the rapids, riffle pool 1, riffle pool 2, andriffle pool 3 sub-reaches of East Creek . . . . . . . . . . . . . 70Figure 5.7 Skew of bed elevation histograms over time in riffles (purpleline) and pools (blue line) in riffle pool 1 in East Creek . . . . 72Figure 5.8 Mean bed deposition and mean bed erosion over time in theriffle units (purple line) and the pool units (blue line) in therapids, riffle pool 1, riffle pool 2, and riffle pool 3 sub-reachesof East Creek . . . . . . . . . . . . . . . . . . . . . . . . . . 74Figure A.1 Distribution of annual bed elevation changes in the riffles andpools within the rapids reach of East Creek from WY04 to WY11 84Figure A.2 Distribution of annual bed elevation changes in the riffles andpools within the riffle pool 1 sub-reach of East Creek fromWY04 to WY11 . . . . . . . . . . . . . . . . . . . . . . . . 85Figure A.3 Distribution of annual bed elevation changes in the riffles andpools within the riffle pool 2 sub-reach of East Creek fromWY04 to WY11 . . . . . . . . . . . . . . . . . . . . . . . . 86Figure A.4 Distribution of annual bed elevation changes in the riffles andpools within the riffle pool 3 sub-reach of East Creek fromWY04 toWY11 . . . . . . . . . . . . . . . . . . . . . . . . . 87Figure A.5 Skew in bed elevation over time in the riffle units (purple line)and the pool units (blue line) in the rapids, riffle pool 1, rifflepool 2, and riffle pool 3 sub-reaches of East Creek . . . . . . 91Figure B.1 Distribution of annual bed erosion and deposition in the rifflesand pools in the rapids reach of East Creek from WY04 to WY11 93Figure B.2 Distribution of annual bed erosion and deposition in the rifflesand pools within the riffle pool 1 sub-reach of East Creek fromWY04 to WY11 . . . . . . . . . . . . . . . . . . . . . . . . 94xivFigure B.3 Distribution of annual bed erosion and deposition in the rifflesand pools within the riffle pool 2 sub-reach of East Creek fromWY04 to WY11 . . . . . . . . . . . . . . . . . . . . . . . . 95Figure B.4 Distribution of annual bed erosion and deposition in the rifflesand pools within the riffle pool 3 sub-reach of East Creek fromWY04 to WY11 . . . . . . . . . . . . . . . . . . . . . . . . 96xvAcknowledgmentsI would like to extend profound thanks to my supervisor Marwan Hassan for hisextensive geomorphology wisdom, enthusiasm, and unmatched patience. I am sovery grateful to have had Marwan as my supervisor. Marwan has managed to bal-ance being a highly prolific and intelligent technical expert and teacher, with alsoextending an incredible degree of genuine compassion and kindness to his students.Marwan’s unwavering belief in me through this process has been invaluable.Thank you very much to Brett Eaton for his technical guidance, support, and forbeing on my supervisory committee. Thank you so very much to Shawn Chartrandfor his technical guidance, on-going support, and perspective - which has beenpivotal to me developing a constructive approach to my thesis and beyond. Manythanks to Shawn for his compassion and for bringing out my rock warrior. Thankyou so much to Joshua Caulkins for his wholehearted support, wealth of knowledgeon East Creek, and superb teaching skills. Josh trained me on East Creek fieldworkand analysis and much of the data used in this thesis was collected, processed, andshared by Joshua Caulkins.Thank you to Marwan Hassan’s students through the years, especially DaveReid, for conducting extensive East Creek fieldwork and contributing to buildingthe database used in this thesis.Thank you very much to my Mom and Dad for their eternal love and supportand for providing me with the foundations to reach this point. Thanks to Momfor caring so deeply and embodying the principles of R.I.C.K. Thanks to Dad forteaching me integrity and the value of the scientific process. Thanks to my sistersTamar and Naomi for their love, support, and advice. Thank you to my UBC Ge-ography, Guelph, and Dundas friends for their love, humour, and encouragement.xviChapter 1Introduction1.1 Background InformationThe study of stream channel stability and sediment storage is critical to our under-standing of stream dynamics. At the broader scale, many resource extraction opera-tions and urban developments depend on some capacity to predict stream behaviourto avoid damage to infrastructure from large sediment mobilizing events. Streamsalso hold ecological value as they provide specialized habitat to fish and otheraquatic organisms. It is, therefore, important that the factors governing streamdynamics are explored and used to inform watershed planning decisions.Knowledge of sediment transport processes in small streams comprises an im-portant contribution to understanding watershed dynamics at a broader scale. Smallstreams are of particular interest within watersheds as they comprise a significantportion of drainage networks on a cumulative length basis (Strahler, 1957; Shreve,1969) and can be more susceptible to small scale disturbances than large streams(Hassan et al., 2005; McCleary and Hassan, 2008). Further, the presence of com-plex morphologies in small steep mountain streams requires a unique approach tocharacterizing these environments. A major factor dictating morphology in moun-tain streams is sediment transport and deposition processes (Hassan et al., 2005).A more complete understanding of sediment transport in small mountain streamshas the potential to substantially aid the development of effective watershed man-agement policies.11.1.1 Channel Stability and AdjustmentStreams are dynamic entities that adjust in response to external environmental con-ditions (Buffington et al., 2003; Hassan et al., 2005). Sediment supply regime andflow regime are two environmental factors that have a major impact on channel dy-namics and morphology (Hassan et al., 2007). Streams tend towards an equilibriumstate when they are subjected to changing external conditions so that energy is ex-pended in an optimal way. There are several approaches to describing how streamsrespond to external conditions (Knighton, 2014). One approach to studying energyexpenditures in streams is to focus on visible evidence of channel adjustments asindicators of stability, including channel gradient, bed material characteristics, andchannel morphology (Church, 2006; Knighton, 2014; Frothingham et al., 2002).Channel gradient is often assessed using longitudinal profiles (Knighton, 2014).Longitudinal profiles reveal changes in the elevation and slope of a channel overtime (Hassan et al., 2005). This can be used to infer about sediment transport pro-cesses, erosion and deposition patterns, and about whether a channel is in a stateof aggradation, degradation, or equilibrium (Lisle et al., 1992; Knighton, 2014;de Almeida and Rodrı´guez, 2011).Bed material characteristics can be assessed by analyzing grain properties andconfiguration. Grain properties include the material, shape, and grain size distri-bution on the channel bed. Configuration includes bedforms and arrangement ofparticles on the bed of a channel. In gravel bed streams, changes to bed configura-tion tend to occur at a spatial scale of 10 to 100 m and at a time scale of around 10 to100 years, roughly consistent with the timescale of interest in this study (Knighton,2014; Hassan et al., 2007).Channel morphology can be described using various classification schemeswith Montgomery and Buffington’s (1997) process based framework being a com-monly referenced system. Channel morphology incorporates channel gradient andbed material characteristics in addition to channel planform, hydrological envi-ronment, sediment sources, and sediment storage elements, among other factors.Channel morphology is important for understanding channel stability because mor-phology adjusts in response to changing flow and sediment supply conditions andalso impacts the rates and spatial distribution of sediment transport processes within2a channel (Buffington and Montgomery, 1997; Hassan et al., 2005).1.1.2 Channel MorphologyMontgomery and Buffington (1997) identify seven distinct channel reach mor-phologies based on slope, grain size, shear stress, and roughness. This paper willfocus on those morphologies present in the East Creek study reach: rapids, rifflepool, and step pool morphologies.RapidsIn this paper Zimmermann and Church’s (2001) ‘rapids’ classification will be usedinterchangeably with Montgomery and Buffington’s (1997) ‘plane bed’ classifi-cation. Rapids morphologies can occur in confined or unconfined channels withmoderate to high gradients of 2 to 10 percent (Buffington and Montgomery, 1997;Hassan et al., 2005; Clifford, 1993). Rapids commonly occur in gravel to cobblebed streams; however, they can also occur in sand bed streams. The source of sed-iment is usually from fluvial sources, bank failure, or debris flow and is stored inoverbank deposits. Rapids are characterized by long nearly featureless stretchesof bed. The primary roughness elements that produce flow resistance in rapidsreaches are the channel banks and grain scale features (Montgomery and Buffin-gton, 1997). Grain scale features in rapids include clusters, stone lines, and stonenets along well-armoured bed surfaces. The characteristically well-armoured bedsof rapids morphologies are suggestive of sediment supply limited conditions; how-ever, transport limited conditions have also been commonly observed in armouredgravel bed channels. This implies that rapids may represent a unique transitionalmorphology between supply and transport limited states (Montgomery and Buffin-gton, 1997; Hassan et al., 2005).Riffle PoolsIn this paper ‘riffle pool’ and ‘pool riffle’ will be used interchangeably to describethe ‘pool-riffle’ morphology presented by Montgomery and Buffington (1997).Riffle pool sequences occur in unconfined channels with moderate to low gradi-ents of around 1 percent (Buffington and Montgomery, 1997; Hassan et al., 2005;3Clifford, 1993). They commonly occur in gravel bed streams; however, grain sizesmay range from sand to cobble. The source of sediment in pool riffle sequences isusually from fluvial sources or from bank failure and is stored in bedforms. Thesebedforms and grain roughness are the dominant roughness elements which produceflow resistance (Montgomery and Buffington, 1997). The riffles in riffle pool se-quences are characterized as flat areas made of gravel, commonly in a lobate shape,and tend to contain coarse tightly interacting particles. In contrast, pools are deepareas with finer material that have looser interactions (Hassan et al., 2005; Clif-ford, 1993; Thompson, 2011). The spacing of riffles and pools has been observedto range from 1.5 to 23.3 channel widths apart, but on average they are typicallyspaced at about 5 to 7 channel widths apart. The range in spacing may partly be at-tributed to the presence of large wood inputs which decrease pool spacing (Hassanet al., 2005; Buffington and Montgomery, 1997).Step PoolsStep pool sequences occur in confined channels with moderate to high gradientsof greater than 3 percent (Buffington and Montgomery, 1997; Hassan et al., 2005).Step pools commonly occur in cobble boulder streams, but require heterogeneousbed mixtures to form. The source of sediment in step pools is usually from fluvialsources, hillslopes, or debris flows and is stored in bedforms. These bedforms,channel banks, and grain roughness are the dominant roughness elements whichproduce flow resistance. The steps in step pool sequences are characterized as dis-crete channel-spanning features associated with an elevation drop (Montgomeryand Buffington, 1997). They can be comprised of large boulders or an accumu-lation of large grains that create local flow resistance and vary in height. Stepsare spaced every one to four channel widths and separate pools (Montgomery andBuffington, 1997; Hassan et al., 2005). In contrast, pools are deep areas with finermaterial that have looser interactions (Hassan et al., 2005; Clifford, 1993; Thomp-son, 2011). Step pools have been proposed to form during sediment supply limitedconditions and offer bed stability (Montgomery and Buffington, 1997).41.1.3 Sediment Transport and StorageSediment transport and changes in sediment supply and discharge dictate the pres-ence of various stream morphologies including rapids, riffle pools, and step pools(Lenzi et al., 1999). Sediment transport mechanisms are the product of interactionsbetween sediment grain properties, bed composition, flow velocity, flow depth,and energy gradient, among other factors (Gomez and Church, 1989; Kondolf andPie´gay, 2003; Wilcock, 2001; Hassan and Reid, 1990). The complex and con-founded nature of such interactions both complicates measurement and modelingefforts and creates great potential for future exploration. To better understand sedi-ment transport and storage patterns, it is important to recognize classifications andmechanisms of sediment transport.Sediment Transport ClassificationSediment transported through water may be divided into different categories de-pending on the classification principle used. The two most commonly used classi-fication principles are: (1) mechanism and (2) morphology.(1) Using mechanism as a classification principle, sediment load can be catego-rized into bed load and suspended load. Suspended load is comprised of fine sedi-ment that moves in suspension in water and is supported by fluid forces; whereas,bed load is comprised of coarse sediment that travels in contact with the bed of thechannel and is supported by both fluid forces and the channel bed (Leopold, 1994;Gomez and Church, 1989). Bed load is collected using a bed load trap sampler. Abed load trap sampler captures the coarser sediment that is transported by rolling,sliding, or saltation.(2) Using morphology as a classification principle, sediment load can be cat-egorized into wash load and bed material load. Wash load can be distinguishedfrom bed material load using the size of the sediment. Wash load consists of thefiner fractions of sediment and bed material load consists of the coarser fractions ofsediment. Bed material load can be measured using digital elevation model analy-ses. Bed material load may contribute to some of the bed load; however, bed loadtypically does not encompass all of the bed material load.This study will focus on the coarser fractions of sediment - the bed load and5bed material load.Bed load Transport MechanismsDepending on sediment supply, discharge, grain size distribution, and grain inter-actions, sediment in a stream may experience various states of mobility (Dietrichet al., 2006; Lisle et al., 2000; Venditti et al., 2010). Venditti et al. (2010) identifythree states of sediment mobility: partial transport, selective transport, and equalmobility. When a channel is in a state of partial transport, all particle sizes aremobilized but more fine material is mobilized compared to that on the bed sur-face (Venditti et al., 2010). In selective transport, only a fraction of sediment sizesare mobilized with the material on the bed surface containing coarser material notfound in the mobilized bed load (Venditti et al., 2010). Which particles are mobi-lized and which particles remain on the bed during a selective transport situationdepends on the size of the particles. As a result of selective transport, sedimentsorting by size may occur and may produce distinct patterns on the channel bed(Wittenberg et al., 2007). In equal mobility, the grain size distribution on the bedsurface is the same as that of the bed load (Venditti et al., 2010; Parker, 2008; Yuillet al., 2010).The relationship between grain size distribution and mobility is complex. In or-der for large grains on the bed of a channel to be moved, the shear stress exerted ona particle by the fluid must be greater than the opposing frictional forces. Once thethreshold of motion is overcome, the bed load becomes mobile and may be trans-ported through sliding, rolling, or saltation processes (Leopold, 1994). However,as implied by the phenomenon of selective transport, grain size is not the sole de-terminant of force required to move a grain. Interlocking of particles, armouring ofcoarser particles on top of finer particles, and spatial variability of bed shear stresscan also impact the conditions required for initiation of motion. For instance, usingflume experiments Nelson et al. (2009) found that sediment flux estimations maybe inaccurate because of variability in shear stress and grain size across the width ofthe channel. To further complicate matters, although Lisle et al. (2000) observedthat variations in boundary shear stress control bed load transport; in an earlierstudy, Garcia et al. (2000) found that there was not a direct relationship between6grain shear stress and bed load flux because of variations in mobility thresholdsand variation in bed characteristics. Oldmeadow and Church (2006) confirmed therelevance of bed characteristics to sediment transport through their findings thatsurface armouring has an observable impact on sediment transport rates and thatgravel bed rivers tend to be in a partial transport regime such that some bed mate-rial may be stationary and other bed material may be mobile. Additionally, Paolaand Seal (1995) proposed that even during equal mobility conditions, selective de-position can still occur which leads to sediment sorting.In addition to grain size distribution and configuration, sediment supply is alsoan important consideration in predicting sediment mobility. Iseya and Ikeda (1987)observed that the availability of sediment particles is a key influence on the mag-nitude and occurrence of bed load transport. Lisle et al.’s (2000) findings that insediment poor channels there are small areas of concentrated high mobility andlarge areas of low mobility and partial mobility supports the observation that sed-iment availability can affect the state of mobility of sediment. Also, Lamarre andRoy (2008) found that sedimentary structures influence sediment transport. Stud-ies on sediment transport in gravel bed rivers have largely focused on discharge atwhich particle entrainment occurs (Wittenberg and Newson, 2005); however, sed-iment transport depends on a complex array of factors including sediment supply,grain size distribution, and grain interactions (Frey and Church, 2011; Yuill et al.,2010).1.1.4 Bed load Storage QuantificationBed load transport and storage is typically quantified by taking measurements inthe field and applying bed load transport formulae to the field data. Common fieldmeasurement techniques include using handheld bed load samplers, pit traps, trac-ers, scour chains, and magnetic detection systems. These techniques offer varyinglevels of cost and accuracy for different time and spatial scales. Similarly, bedload transport formulae range in level of accuracy. Bed load transport formulaeare typically a function of some combination of water discharge, velocity, watersurface slope, grain density, water density, grain size, and grain shape. Commonbed load transport formulae include the DuBoy, Shields, Meyer-Peter and Muller,7Bagnold, and Wilcock and Crowe equations. These formulae can offer useful in-sights into bed load transport processes; however, they are not applicable for allcontexts and without careful consideration of the characteristics of the watershedin question they can be used inappropriately. Even when conscientiously applied,sediment transport functions often vary from measured rates by more than an orderof magnitude (Hassan et al., 2007). Consequently, it could be worthwhile to ex-plore non-classical methods of predicting sediment transport and storage patterns.Church (2006) recommends using an inverse approach to investigate sedimenttransport and storage. In the inverse approach, the morphological properties of astream are used as a starting point to infer about sediment transport processes oc-curring through riffles and pools. Given that bed load transport is a key predictorof sediment balance and channel morphology, analysing a sediment budget in con-nection with changes in channel form can, in turn, be telling of sediment transportrates (Church, 2006). There are many examples of applying the inverse approachfor braided sand bed rivers; however, despite it’s unique ability to capture vari-ability in sediment transport, it has received only modest attention in gravel bedstreams (Ham and Church, 2000).1.2 Research Gap and Study ObjectivesIn spite of research previously conducted on sediment transport, we lack basic un-derstanding of sediment storage, supply, morphology, and flow regime. To betterunderstand sediment transport processes, it is important to know how channel mor-phology responds to changes in flow and sediment supply. This thesis will utilizedata from East Creek, a small mountain stream in Coastal British Columbia, to dis-cuss spatial and temporal patterns of sediment mobility and storage. The objectivesof this study are to address the questions:1. How does annual sediment storage relate to channel morphology at the chan-nel, reach, and unit spatial scales in East Creek?2. How does reach-scale sediment storage relate to annual peak discharge inEast Creek?8Chapter 2Study SiteThe study was conducted in a small second-order gravel bed mountain stream lo-cated in the UBC Malcolm Knapp Research Forest in the Fraser Valley foothills ofthe Coast Mountains in British Columbia, Canada (Figure 2.1). The East Creekwatershed is approximately 100 ha and the area receives between 2000 and 2500mm of mean annual precipitation. The upper portion of the East Creek study areawhere the study site is located is dominated by young Douglas-fir trees (Pseu-dotrsuga menziesii), Red Alder (Alnus rubra), and Salmonberry bushes (Rubusspectabilis). The channel ranges in width from 2 to 5 m and has an average gradi-ent of 3% (Caulkins).9Figure 2.1: Location of UBC Malcolm Knapp Research Forest in Fraser Val-ley Foothills, British Columbia. The research forest is outlined in pur-ple. (adapted from Google Maps)The study reach extends approximately 600 m in length and encompasses threedistinct morphologies: rapids, riffle pool, and step pool (Figure 2.2). The entirestudy channel is bounded by a culvert at the upstream end of the top of the rapidsand by a pit trap at the downstream end of the step pool. A road named ”M road”lies between the riffle pool and step pool sub-reaches, which are, consequently,connected via culvert. In this thesis, at the channel scale, analyses are presentedthat extend from the upstream end of the top of rapids reach to the downstreamend of the step pool reach. At the reach and unit scales, additional, more detailedanalyses are presented for the rapids and riffle pool morphologies.10Figure 2.2: Location of rapids, riffle pool 1, riffle pool 2, riffle pool 3, andstep pool in East Creek, British Columbia2.1 Top of Rapids“Top of rapids” refers to a short (5.9 m) length of channel upstream of the rapidsreach, bounded by a culvert at the upstream end and a pit trap at the downstreamend. The top of rapids (TOR) exhibits similar morphology to the rapids reach butcontains a distinct plunge pool created by the culvert discharge. The plunge poolspans nearly half of the TOR and has a substantial impact on sediment transportand storage in this reach. For the purposes of the sediment budget element of thisstudy, the TOR reach is effectively treated as a buffer zone so as not to confoundthe relationship between sediment storage and morphology with culvert inducedimpacts on sediment storage.2.2 RapidsThe rapids reach is 84 metres in length and lies between the top of rapids and rifflepool reaches. It is bounded by pit traps at its upstream and downstream ends. Arelatively shallow gradient (2.7%) and coarse bed material characterize the rapidsreach, with a surface D50 of 57 mm and a subsurface D50 of 31 mm. Steep over-hanging banks, patches of exposed till, and discontinuous sediment structures in-cluding stone lines and stone cells can be observed in the rapids (RAP).112.3 Riffle PoolThe riffle pool reach (381 m) is significantly longer than the rapids (84 m) and steppool (approx. 120 m) reaches. To improve comparability with the other reaches,the riffle pool reach was divided into three smaller sub-reaches: riffle pool 1 (121m), riffle pool 2 (163 m), and riffle pool 3 (97 m). These sub-reaches have shallowgradients of 1.8%, 1.5%, and 0.9%, respectively, and exhibit an intermediate grainsize distribution as compared to the rapids and step pool reaches.The riffle pool 1 (RP1) sub-reach is bounded by pit traps at the upstream anddownstream ends (Figure 2.3). The sediment observed in RP1 is finer than that ofthe rapids reach directly upstream and is coarser than that of the riffle pool 2 reachdirectly downstream, with a surface D50 of 42 mm and a subsurface D50 of 21 mm(Caulkins). Like RP1, the riffle pool 2 (RP2) sub-reach is bounded by pit traps atthe upstream and downstream ends. The sediment observed in RP2 has a surfaceD50 of 32 mm and a subsurface D50 of 14 mm (Caulkins). The riffle pool 3 (RP3)sub-reach is bounded by a pit trap at the upstream end and by a culvert that runsunder M Road at the downstream end. Grain size distribution data for riffle pool3 is not available. Given that there is no pit trap at the downstream end of RP3,comparatively more attention is given to RP1 and RP2 in the sediment budget. Thethree riffle pool sub-reaches are collectively characterized by heterogeneous bedmaterial with bed structures including alternating sequences of riffles, pools, runs,and bars (Figure 2.4).12Credit: Joshua CaulkinsFigure 2.3: Photograph of a pit trap outlined in red, located at the boundarybetween the rapids and riffle pool 1 in East Creek, British Columbia13Credit: Tony Lagemaat and Dave ReidFigure 2.4: Aerial photograph of riffle pool 1 sub-reach in East Creek takenusing pole photography method2.4 Step PoolThe step pool reach roughly spans a length of 120 metres and is the most down-stream reach of the study area. The step pool (SP) reach is bounded by a culvertthat runs under M Road at the upstream end and by a pit trap at the downstreamend. The grain size distribution in the step-pool reach is visibly more coarse than inthe other reaches; however, complications due to sediment mixing prevent accuratesampling of particle size in this area. The step pool reach has the steepest gradi-ent (8.8%) and contains steps formed from woody debris and heterogeneous grainmixtures (Caulkins). The sediment budget element of this study does not includethe step pool reach; however, the step pool is included in channel scale analyses.14Chapter 3Methods3.1 Data CollectionThere is an extensive dataset on channel morphology, sediment transport, and dis-charge in East Creek that extends from 2003 to 2011. This dataset has been col-lected, compiled, and analyzed by a number of different people over the yearsunder the supervision and direction of Marwan Hassan. I participated in the lateryears of the data collection towards building this database; however, the majorityof the data used in this thesis comes from the existing data set and builds on thedata processing and analyses of other students and researchers. The remainder ofthis section will focus on the methods that were used by a number of individuals tocollectively gather the existing data and the methods that I used to analyze the dataprovided to me.3.1.1 Pit TrapsPit traps were used in East Creek as a measure of discrete event-scale sediment stor-age. Sediment deposited in the traps was collected and weighed after each stormevent. Storm events occurred during the winter and early spring. The impact ofthe traps to sediment transport dynamics was minimized by returning all collectedsediment to the channel just below the trap immediately after recording its weight.This facilitated the re-entry of the material into the stream system and returned thetrap to an empty state to allow for sediment capture in the next storm event. In15this way, the impact of the temporary removal of sediment from the system via thetraps was limited to the event scale, with no seasonal or annual impacts.There were five wooden pit traps in total, each of which spanned the widthof the channel. They were located at the upper bound of the rapids, the rapids-riffle pool 1 interface, the riffle pool 1-riffle pool 2 interface, the riffle pool 2-rifflepool 3 interface, and at the lower bound of the step pool. There was no pit traplocated at the interface between riffle pool 3 and the step pool. This boundary wasdemarcated by M Road. Each pit trap was dug into the bed of the channel to adepth of 0.24 m - 0.3 m, such that the top of each trap lay flush with the surface ofthe channel bed (Caulkins). The pit trap volumes and estimated sediment trappingcapacities are given in Table 3.1.Table 3.1: Pit trap volumes and estimated sediment trapping capacitiesLocation Volume Estimated Capacity(m3) (kg)Rapids Upper Bound 1.07 1853Riffle Pool 1 Upper Bound 0.27 473Riffle Pool 2 Upper Bound 0.67 1162Riffle Pool 3 Upper Bound 0.30 515Step Pool Lower Bound 0.63 10933.1.2 Ground SurveysAnnual mapping of the channel topography of East Creek from 2003 to 2011 wascarried out using a theodolite-based total station equipped with an electronic dis-tance meter (EDM) (Caulkins). Operation of the total station involved one personfocusing an eyepiece within the total station on an optical prism held by a secondperson at a predetermined location. With the eyepiece focused on the prism, the to-tal station operator prompted the emission of a laser beam, which hit the prism andwas reflected back to the station. Using the theodolite and EDM, the total stationinternally calculated the horizontal angle, vertical angle, and distance to the prismcreating X, Y, and Z coordinates for the data point.16Data points were collected at 0.5 m intervals across the width of the channelbed and at 0.5 m intervals along the length of the channel bed for the 600 m reach,creating a 0.5 by 0.5 m grid of elevations, corresponding to a point density of 9points per square meter. Channel banks were also surveyed at 0.5 m intervals to adistance of approximately 1 m out from the edge of the bed on either side of thechannel. Unique characteristics such as bed, banks, woody debris, till, and islandswere distinguished from each other by entering codes on the keypad of the totalstation. Boulders and large stationary rocks distributed throughout the channelwere spray painted and also were surveyed to be used as control points for laterphotograph rectification.The locations of seventy-four sets of stationary rebar pins spaced along the leftand right channel banks at 5 to 15 m intervals had been established in a previousyear of the study (Caulkins). Some of these pins were re-surveyed and comparisonof surveyed pin locations to known pin locations were used to estimate and correctsurveying errors.3.1.3 Aerial PhotographyAerial photographs of the channel were taken to enable channel mapping. A cam-era suspended atop a 10 m metal pole was raised in the air by field personnelto obtain an aerial frame of the creek. The pole and camera were carried down-stream along the length of the creek with photographs taken by remote control atan overlap of approximately one half to one third of a frame between successivephotographs. These photographs were taken annually at the end of the summercorresponding to low flow conditions for optimal visibility of bed features.3.2 Data AnalysisAs with the field data collection, data cleaning and analysis had been conducted onmuch of the field data prior to this study. I have built on the analyses performed byothers by conducting the analyses described below.173.2.1 Storage Estimates using Pit TrapsBed load flux estimates between the 2008 and 2010 water years were generatedfor each reach using data from the pit traps and aluminum traps located at thedownstream boundary of each reach. From this point onwards, water year (WY) isdefined as the period from October 1 in the prior year to September 30 of the givenyear. For example, WY09 refers to the period from October 1 2008 to September30 2009 and WY10 refers to the period from October 1 2009 to September 2010.Since some reaches were bounded by both a pit trap and an aluminum trapand others were bounded only by a pit trap, two different measures of volumeof sediment exported were calculated. Bed load export values as calculated fromthe pit traps alone are presented for all reaches to allow for greater consistencyin comparisons across reaches. In reaches where there was both an aluminumtrap and a pit trap, bed load export values taken as the sum of bed load retrievedfrom both traps combined is presented. The combined results were used to createthe sediment budget because the coarser sediment fractions that were trapped inthe aluminum traps would have been difficult to re-mobilize once trapped, andtherefore, once in the trap this material was effectively, temporarily, removed fromthe stream. Bed load transport rates were not calculated for the riffle pool 3 sub-reach because there was no pit trap at the downstream end of this sub-reach.Bed load storage was calculated using equation 3.1.∆SB = Si−So (3.1)where Si is bed load input into a reach as given by the mass of sediment in thetrap at the upstream bound of the reach and So is bed load output from a reach asgiven by the mass of sediment in the trap at the downstream bound of the reach.3.2.2 Storage Estimates using Ground SurveysBed material erosion and storage estimates were generated for the rapids reach, andthe riffle pool 1, 2, and 3 sub-reaches of East Creek from WY04-11 using elevationchange mapping data.Surveyed elevation data was imported into ArcGIS and used to create shape-files that displayed the annual surveyed data on a point by point basis; triangulated18irregular networks (TINs) of the study reach to annually interpolate elevation val-ues between surveyed points; and annual digital elevation models (DEMs) to createsmooth elevation surfaces that could be compared for consecutive water years. Al-though bank points were surveyed, they were removed for the analysis of erosionand deposition patterns because there was a high degree of uncertainty associatedwith elevation change estimates along the banks. This high uncertainty can be at-tributed to the comparatively lower point density and sharper topographic changesaround the banks.Difference maps were created from the DEMs using A MATLAB program cre-ated by Joseph Wheaton to quantify annual bed elevation changes and the level ofuncertainty associated with those changes (Wheaton et al., 2010). A brief descrip-tion of the program follows; however, Wheaton’s (2008) thesis provides a far morecomprehensive description of the program. Wheaton describes multiple pathwaysthat may be used in this program depending on the objectives of the study. Thisstudy used pathway 4 (Wheaton, 2008).The difference maps (DoDs) were created by subtracting the elevations of oneyear from the elevations of the following year on a cell by cell basis. Similarly, theuncertainties associated with elevation estimates were calculated on a cell by cellbasis. The uncertainties were determined using a fuzzy inference system whichused multiple qualitative criteria to create a final quantitative estimate of the un-certainty (Jang and Gulley, 1995). The qualitative criteria used in this study werepoint density and slope, where a high point density corresponded to a low level ofuncertainty and a high slope corresponded to a high level of uncertainty. The qual-ity of the point density and slope were represented by categories of low, medium,and high. The category of point density and slope for each cell together were usedto determine an uncertainty estimate for that cell. The point density and slopeinputs into the program were created in ArcGIS and were derived from the orig-inal surveyed bed elevation data. A confidence interval of 95% was used as athreshold to propagate the calculated uncertainties onto the map. The gross eleva-tion change estimates and the uncertainty-adjusted elevation change estimates wereused to create two sets of difference maps for 2003 to 2010. The MATLAB pro-gram output also included numerical distributions of the gross elevation changesand the uncertainty-adjusted elevation changes for each pair of years.19Annual bed material erosion volumes obtained from the difference maps wereconverted into annual bed material erosion masses using equation 3.2 and using theconstants indicated in Table 3.2.Qm =VeφρLt/Lrt(3.2)in which Ve is volumetric bed material erosion , φ is the porosity of the bed material,ρ is the density of the bed material approximated by the density of granite, Lt isthe distance of travel of mobilized bed material approximated by the average steplength obtained from tracer stones, Lr is the distance over which Ve is determinedwhich is equivalent to the reach length, and t is the time interval between surveys.Table 3.2: Values of constants used to calculate bed material erosion massesConstant Value UnitsPorosity of Sediment (φ ): 0.25Density of Granite (ρ): 2600 kg/m3Average Step Length (Lt): 7.56 mRapids Length (Lr1): 84.2 mRiffle Pool 1 Length (Lr2): 120.6 mRiffle Pool 2 Length (Lr3): 163.4 mRiffle Pool 3 Length (Lr4): 97 mBed material storage was calculated using equation 3.3.∆SM = Sd−Se (3.3)where Sd is bed material deposition within a reach as calculated from digital el-evation difference maps and Se is bed material erosion within a reach as calculatedfrom digital elevation difference maps.In addition to using the survey data to create a bed material budget, the sur-vey data was also used to create longitudinal profiles. A grid subtraction betweenDEMs for 2003 and 2009 was conducted using GIS to calculate the elevation dif-ferences between 2003 and 2009 on a cell by cell basis. A line running through thecentre of the channel was created as a surrogate for the channel thalweg. Elevationsand downstream distances were recorded at the intersection of this centreline and20each cross section to create longitudinal profiles of the channel.3.2.3 Morphology Mapping using Aerial PhotographyAerial photographs of the channel were spatially rectified in ArcGIS and used togenerate a map of morphological features in the channel. Each annual set of aerialphotographs was rectified using surveyed points that included large stationary con-trol point boulders, rebar pins, and trap boundaries. The morphology map usedin this study was created by Caulkins (personal communication, 2010) based onthe morphological features visible in the rectified 2005 aerial photographs and anintimate knowledge of the channel acquired from multiple years of fieldwork ex-perience at the East Creek study reach. The 2005 morphology map was used to de-marcate morphological features for all years within the study period to more easilyaccommodate comparisons over time. The potential for error that could arise fromextrapolating morphological feature locations over multiple years was gauged byoverlaying the morphology map generated using the 2005 aerial photographs onmaps of rectified aerial photographs for each year of the study period. There werelittle to no changes observed between the location and extent of the features demar-cated in the 2005 morphology map and those observed in the rectified photographsfor the remaining years, suggesting that there would have been only minimal errorsassociated with using the 2005 map for all years, given the scale and scope of thisstudy.21Chapter 4ResultsThe results section presents sediment storage and transport trends observed in EastCreek organized using spatial scale. First, flow regime data for East Creek areprovided. Next, sediment storage and transport trends at the largest scale analyzedin this study, the channel scale, are presented. Reach scale results follow, allowingfor comparison across the distinct channel morphologies: rapids and riffle pool.Third, unit scale results, the smallest scale analyzed in this study, are reported,allowing for comparison across pools and riffles within the rapids and riffle poolreaches.4.1 Flow RegimeAnnual peak discharge was used as an indication of flow regime in East Creek.Figure 4.1 shows the fluctuations in peak discharge for WY04-11. The highestpeak discharges occurred in WY09 (4.5 m3/s) and WY07 (4.3 m3/s) and the lowestpeak discharge occurred in WY06 (1.0 m3/s).220.0000.5001.0001.5002.0002.5003.0003.5004.0004.5005.00004 05 06 07 08 09 10 11Discharge (m3 /s) Water Year Figure 4.1: Annual peak discharge in East Creek for WY04-114.2 Channel Scale: Trends in Sediment StorageAt the channel scale, sediment storage will be assessed using longitudinal profiles.Longitudinal profiles allow for inferences to be made about the vertical stabilityof a channel because they provide one dimensional estimates of changes in sedi-ment storage and longitudinal adjustments to changes in sediment supply and flowregimes.Figure 4.2 presents the longitudinal adjustment of the channel to flow andsediment supply at the channel scale for the rapids and riffle pool reaches of EastCreek between 2003 and 2009. At this coarse scale, the annual profiles are verysimilar and show little change over time.231301311321331341351361371381391400 100 200 300 400 500Elevation (m) Distance from Cross Section 1 (m) 2003200420052006200720082009Figure 4.2: Longitudinal profile at the channel scale for the rapids and rifflepool reaches of East Creek for 2003 to 2009Longitudinal profiles are also presented separately, at a more resolved level, forthe rapids, riffle pool 1, riffle pool 2, and riffle pool 3 to discern changes at the finescale that cannot be observed at the coarse scale (Figure 4.3).24136.5137137.5138138.51390 20 40 60 80 100Elevation (m) Distance from Cross Section 1 (m) (a) Rapids134134.5135135.5136136.513770 90 110 130 150 170 190 210 230Elevation (m) Distance from Cross Section 1 (m) (b) Riffle Pool 1132132.5133133.5134134.5135190 210 230 250 270 290 310 330 350 370Elevation (m) Distance from Cross Section 1 (m) (c) Riffle Pool 2131.2131.3131.4131.5131.6131.7131.8131.9132132.1132.2132.3350 370 390 410 430 450 470Elevation (m) Distance from Cross Section 1 (m) (d) Riffle Pool 3Figure 4.3: Longitudinal profiles at the reach scale for the rapids and riffle pool reaches of East Creek for 2003 to 200925Locations within the channel where the greatest range in elevation change oc-curred are identified as areas of interest. The morphology, width, and channel pat-tern at these locations are summarized in Table 4.1. The presence of large woodand its upstream (US) or downstream (DS) location, corresponding to the distanceson the longitudinal profiles are also noted. The most substantial elevation changeover the six year period corresponded to the upper part of the rapids reach, from9.2 m to 24.2 m downstream. The average bed elevation change over this area was0.36 m. The channel width in this section was measured at the section midpoint of15.2 m downstream. For all other areas in the channel highlighted in Table 4.1,substantial bed elevation change was more localized around a shorter length ofchannel. In these cases, the reported distance downstream and channel width weremeasured directly at the noted distance downstream.26Table 4.1: Morphology at areas of interest based on longitudinal profiles of East Creek upper reaches between 2003and 2009Distance Range of Elevation Reach Morphological Channel Channel NotesDownstream Change Unit Width Pattern(m) (m) (m)2.9 - 24.2 0.36 rapid run 3 straight DS of plunge pool65.5 0.24 rapid pool 3.4 very wide bend DS of LW146.2 0.59 riffle pool 1 side bar 4.9 sharp bend LW intersects XS166.7 0.24 riffle pool 1 pool 3.3 wide bend –270.9 0.31 riffle pool 2 pool 1.9 very wide bend DS of LW; US of LW and back channel313 0.28 riffle pool 2 side bar 6.2 straight US of LW and back channel373.8 0.26 riffle pool 2 run 2.8 straight US of pit trap27In all reaches, the annual elevation profiles are fairly consistent. Of the fewareas that experienced notable changes in bed elevation, these changes fluctuatedbetween aggradation and degradation over the six year period.Figure 4.4 shows the change in channel gradient in each of the upper reachesover the 2003 to 2009 period. The rapids experienced the largest change in channelslope (0.0008 m/m) over time compared to the relatively stable slopes observed inthe riffle pool sub-reaches (all ≤0.0002 m/m). The greatest annual changes inchannel slope (calculated as the percentage difference from the average slope inthat sub-reach) occurred in the rapids in WY07 (14%), in riffle pool 1 in WY07(3%), in riffle pool 2 in WY05 (8%), and in riffle pool 3 in WY06 (11%) closelyfollowed by WY07 (9%).y = - 0.0008x + 1.5775  y = 0.0001x -  0.2387 y = - 0.0002x + 0.3653  y = 4E - 05x -  0.0765 00.0050.010.0150.020.0250.032002 2003 2004 2005 2006 2007 2008 2009 2010Channel Gradient (m/m) Year RA PRP1RP2RP3Figure 4.4: Change in channel gradient over time across the upper reaches ofEast Creek from 2003 to 2009284.3 Reach Scale: Sediment Flux and Storage4.3.1 Bed load Transport from Pit TrapsBed load flux and storage estimates were generated for the upper sub-reaches ofEast Creek for WY09-11 using sediment trap data. Bed load data is also availablefor East Creek for additional years. It was selected to limit the focus of the bed loadflux component of this study to the water years listed above in favour of conductinga more comprehensive analysis of bed material transfer (Subsection 4.3.2) whilemaintaining a feasible study scope .Bed load FluxTable 4.2 presents bed load flux estimates from the pit traps alone and from the pitand aluminum traps combined for WY09-11 for the upper reaches of East Creek.The mass of bed load transported ranged from 431 kg to 2576 kg. The averagemass of bed load transported per reach per year was 1102 kg. Of the reacheslisted in Table 4.2, the top of rapids reach experienced the largest annual bed loadtransport fluxes averaging 1547 kg/yr. Of the remaining reaches, the rapids reachexperienced the largest annual bed load transport flux averaging at 1295 kg/yr andthe riffle pool 2 sub-reach experienced the smallest annual bed load transport fluxaveraging at 732 kg/yr.Bed load StorageAnnual bed load storage estimates are presented in Table 4.3 for the rapids, rifflepool 1, and riffle pool 2 sub-reaches for WY09-11. Storage estimates for the top ofrapids and riffle pool 3 are not given because the top of rapids reach did not have apit trap on its upper bound and the riffle pool 3 sub-reach did not have a pit trap onits lower bound.Over the three year period, the rapids reach exhibited the largest storage (1252kg), and riffle pool 2 exhibited the smallest storage (29 kg). Of the nine bed loadstorage estimates obtained from the pit traps, seven demonstrate net deposition asindicated by positive storage values and two demonstrate net erosion as indicatedby negative values. The cases of net erosion occurred in the rapids reach in WY1129Table 4.2: Bed load flux estimates using pit traps for WY09-11Reach Water Year Pit Trap Flux Combined Pit andAluminum Trap Flux(kg) (kg)TOR WY09 992 n/aTOR WY10 2576 n/aTOR WY11 1074* n/aRAP WY09 768 836RAP WY10 1224* 1324*RAP WY11 1603 1725*RP1 WY09 560 573RP1 WY10 400 431RP1 WY11 1152 1193*RP2 WY09 544* n/aRP2 WY10 1280* n/aRP2 WY11 1046* n/aSP WY09 992 1019SP WY10 552 598SP WY11 1290 1315**A complete record of bed load mobilizing events could not be ob-tained because of either equipment malfunctioning or overfull trapevents. n/a refers to no aluminum trap present at the bottom of thereach.and in the riffle pool 2 sub-reach in WY10.4.3.2 Bed Material Transfer from Morphological MethodBed material erosion and storage estimates were generated for the rapids reach,and the riffle pool 1, 2, and 3 sub-reaches, of East Creek from WY04-11 usingelevation change mapping data.Bed Material ErosionTable ( 4.4) shows that the annual mass of bed material transported ranged from 57kg to 767 kg, where erosion mass was derived from erosion volume using equation30Table 4.3: Bed load storage based on pit traps for WY09-11Station Water Year Bed load Storage(kg)RAP WY09 156RAP WY10 1252RAP WY11 -651RP1 WY09 263RP1 WY10 893RP1 WY11 532RP2 WY09 29RP2 WY10 -849RP2 WY11 147( 3.2). The average mass of bed material transported per reach per year was 218kg. Listed in descending order, the average bed material transport fluxes over theeight year period for the rapids, riffle pool 2, riffle pool 1, and riffle pool 3 are 254kg/yr, 245 kg/yr, 197 kg/yr, and 176 kg/yr.Bed Material StorageBed material storage estimates were derived using summed differences from digitalelevation models. The differences between deposition and erosion as indicated byEquation 3.3 are shown for WY04-11 in Figure 4.5 for the rapids and riffle pools1, 2, and 3.Bed material storage estimates were used to analyze temporal patterns withineach morphology. Within the rapids reach, the largest (-9.67 m3) and smallest (-0.05 m3) magnitudes of net change occurred in WY07 and WY11, respectively(Figure 4.5a). Sediment storage fluctuated over the eight year study period withlarger magnitude changes more concentrated in the first half of the study period.Within the riffle pool 1 sub-reach, the largest (-3.62 m3) and smallest (0.03m3) magnitudes of net change occurred in WY07 and WY08, respectively (Figure4.5b). Sediment storage fluctuated over the eight year study period with largermagnitude changes more concentrated in the first half of the study period.31Table 4.4: Bed material export estimates using morphological methods forWY04-11Reach Water Year Bed Material Bed MaterialErosion Volume Erosion Mass (Se)(m3) (kg)RAP WY04 2.1 124WY05 8.1 474WY06 2.7 156WY07 13.1 767WY08 2.1 121WY09 1.6 96WY10 2.5 148WY11 2.5 147RP1 WY04 4.0 163WY05 7.9 322WY06 5.7 233WY07 9.3 377WY08 3.8 153WY09 3.3 133WY10 1.7 70WY11 3.0 123RP2 WY04 7.3 220WY05 15.1 455WY06 4.6 139WY07 15.4 462WY08 5.0 151WY09 4.8 144WY10 7.6 229WY11 5.3 160RP3 WY04 4.9 248WY05 5.3 270WY06 3.3 168WY07 3.9 200WY08 4.3 217WY09 2.0 100WY10 2.9 147WY11 1.1 5732Within the riffle pool 2 sub-reach, the largest (-3.17 m3) and smallest (0.02m3) magnitudes of net change occurred in WY10 and WY06, respectively (Figure4.5c). Sediment storage fluctuated over the eight year study period with largermagnitude changes distributed throughout the study period.Within the riffle pool 3 sub-reach, the largest (5.11 m3) and smallest (1.3m3) magnitudes of net change occurred in WY11 and WY09, respectively (Fig-ure 4.5d). Sediment storage fluctuated over the eight year study period with adistinct and repeated alternation between erosional and depositional environmentsbetween consecutive years.33-10-8-6-4-2024681004 05 06 07 08 09 10 11Sediment Storage (m3 ) Water Year (a) Rapids-10-8-6-4-2024681004 05 06 07 08 09 10 11Sediment Storage (m3 ) Water Year (b) Riffle Pool 1-10-8-6-4-2024681004 05 06 07 08 09 10 11Sediment Storage (m3 ) Water Year (c) Riffle Pool 2-10-8-6-4-2024681004 05 06 07 08 09 10 11Sediment Storage (m3 ) Water Year (d) Riffle Pool 3Figure 4.5: Volumetric change in sediment storage in the rapids reach and riffle pool sub-reaches of East Creek forWY04-1134Spatial patterns of sediment storage moving downstream along the length ofthe channel are shown separately for each water year in Figures 4.6 and 4.7 forWY04-7 and WY08-11, respectively.In WY04 net storage decreased moving downstream along the channel with de-position in the rapids reach and erosion in the riffle pool 3 sub-reach. In WY07 andWY11, net storage increased moving downstream along the channel with erosionin the rapids reach and deposition in the riffle pool 3 sub-reach. In all other years,net storage fluctuated between aggradation and degradation moving downstreamalong the channel.35-2-1.5-1-0.500.511.522.53RAP RP1 RP2 RP3Sediment Storage (m3 ) Sub - reach (a)WY04-6-5-4-3-2-1012RAP RP1 RP2 RP3Sediment Storage (m3 ) Sub - reach (b)WY05-4-3.5-3-2.5-2-1.5-1-0.50 RAP RP1 RP2 RP3Sediment Storage (m3 ) Sub - reach (c)WY06-12-10-8-6-4-20246RAP RP1 RP2 RP3Sediment Storage (m3 ) Sub - reach (d)WY07Figure 4.6: Volumetric change in sediment storage moving downstream along the bed of the East Creek study reachfor WY04-0736-2-1.8-1.6-1.4-1.2-1-0.8-0.6-0.4-0.200.2RAP RP1 RP2 RP3Sediment Storage (m3 ) Sub - reach (a)WY0800.20.40.60.811.21.4RAP RP1 RP2 RP3Sediment Storage (m3 ) Sub - reach (b)WY09-4-3-2-1012RAP RP1 RP2 RP3Sediment Storage (m3 ) Sub - reach (c)WY10-10123456RAP RP1 RP2 RP3Sediment Storage (m3 ) Sub - reach (d)WY11Figure 4.7: Volumetric change in sediment storage moving downstream along the bed of the East Creek study reachfor WY08-11374.4 Unit-Scale: Distribution of Sediment Storage4.4.1 Bed Elevation Change HistogramsSediment storage and mobility was analyzed separately for the riffle and pool unitsin East Creek, and reported for the rapids reach and the riffle pool 1, 2, and 3sub-reaches.RifflesNet bed elevation change histograms for the riffles were generally uni-modal withonly a few cases of multi-modal histograms. These consisted of one bimodal his-togram in the rapids in WY07; two multi-modal histograms in riffle pool 2 inWY05 and WY07; and two bimodal histograms in riffle pool 3 in WY06 andWY07. Figure 4.8 shows examples of a uni-modal histogram typical of mostyears and sub-reaches ( 4.8a), a rare bimodal histogram ( 4.8b), and a very raremultimodal histogram ( 4.8c) for the riffles. Net bed elevation change histogramsfor all reaches and all years can be found in Appendix B in Figures A.1, A.2, A.3,and A.4 for the rapids, riffle pool 1, riffle pool 2, and riffle pool 3, respectively.(a) A Uni-modal his-togram in rapids for riffles inWY04(b) A Bi-modal his-togram in riffle pool 3 forriffles in WY07(c) A Multimodal his-togram in riffle pool 2 forriffles in WY05Figure 4.8: Examples of types of bed elevation change histograms in rifflesin East CreekThe means and medians of net bed elevation change in the riffles were close tozero in all years in all sub-reaches, with an average mean of 0.00 m and an averagemedian of 0.00 m. The means ranged from -0.04 m to 0.03 m and the medians38ranged from -0.02 m to 0.03 m. Rounded to the nearest tenth of a decimal, therange of means was equal to the range of medians in riffle pool 2 and riffle pool 3.The average variance in the riffles was 0.005 m2. The largest variance in eachreach usually occurred during WY07 with the exception of the rapids reach wherethe largest variance occurred in WY05. The variance ranged from 0.001 m2 to0.022 m2.The majority of histograms were slightly skewed and there were fluctuationsbetween positively and negatively skewed histograms. The most extreme negativeskews occurred in WY05 in the rapids, riffle pool 1, and riffle pool 3. The mostextreme positive skews were not consistent with the water year. The average skewwas -0.081 and the skews ranged from -4.76 to 3.66. The former occurred in rifflepool 2 and the latter occurred in the rapids.There was a wide range in kurtosis values of the histograms, with an averagekurtosis of 10.0. The minimum (0.37) and maximum (40.25) kurtosis values bothoccurred in riffle pool 2.The summary statistics for riffles in riffle pool 3 are shown as representative forthe other sub-reaches (Table 4.5). Summary statistics for the remaining reachescan be found in Appendix B in Tables A.1, A.2, A.3 for the rapids, riffle pool 1,and riffle pool 2, respectively.39Table 4.5: Summary statistics for bed elevation change in riffles in riffle pool3 sub-reachYear Mean Median Variance Skew Kurtosis(m) (m) (m2)WY04 0.00 -0.01 0.00 0.93 7.79WY05 -0.01 0.00 0.00 -1.20 5.57WY06 0.01 0.00 0.00 1.27 6.70WY07 0.01 0.00 0.00 0.77 1.56WY08 -0.01 0.00 0.00 -1.18 11.18WY09 0.00 0.01 0.00 -0.49 6.29WY10 -0.01 -0.01 0.00 -0.81 4.35WY11 0.01 0.01 0.00 1.16 8.52Average 0.00 0.00 0.00 0.06 6.49PoolsNet bed elevation change histograms for the pools were generally uni-modal withonly a few cases of multi-modal histograms. These consisted of three multi-modalhistograms in the rapids in WY06, WY07, and WY09 and two multi-modal his-tograms in riffle pool 2 in WY05 and WY07. Figure 4.9 shows examples of atypical uni-modal histogram ( 4.9a), a rare bimodal histogram ( 4.9b), and a raremultimodal histogram ( 4.9c) for the pools.40(a) A Uni-modal his-togram in riffle pool 1 forpools in WY09(b) A Bi-modal his-togram in rapids for poolsin WY06(c) A Multimodal his-togram in riffle pool 2 forpools in WY07Figure 4.9: Examples of types of bed elevation change histograms in pools inEast CreekThe means and medians of net bed elevation change in the riffles were close tozero in all years in all sub-reaches, with an average mean of 0.00 m and an averagemedian of 0.00 m. The means ranged from -0.063 m to 0.055 m and the mediansranged from -0.046 m to 0.051 m. Rounded to the nearest tenth of a decimal, therange of means was equal to the range of medians in the rapids.The average variance in the pools was 0.008 m2. The largest variances occurredin WY05 and WY07. The variance ranged from 0.002 m2 to 0.023 m2.The majority of histograms were slightly skewed and there were fluctuationsbetween positively and negatively skewed histograms. The most extreme negativeskews occurred in WY10 in the rapids, riffle pool 2, and riffle pool 3. The mostextreme positive skews occurred in WY11 in riffle pool 1, riffle pool 2, and rifflepool 3. The average skew was 0.015 and the skews ranged from -1.60 to 2.25. Theformer occurred in riffle pool 3 and the latter occurred in riffle pool 1.The kurtosis values were lower in pools than in riffles, with an average kurto-sis of 5.99 in pools. As with riffles, the minimum (0.23) and maximum (14.14)kurtosis values in pools both occurred in riffle pool 2.The summary statistics for pools in riffle pool 3 are shown as representative forthe other sub-reaches (Table 4.6). Summary statistics for the remaining reachescan be found in Appendix B in Tables A.4, A.5, A.6 for the rapids, riffle pool 1,and riffle pool 2, respectively.41Table 4.6: Summary statistics for bed elevation change in pools in riffle pool3 sub-reachYear Mean Median Variance Skew Kurtosis(m) (m) (m2)WY04 -0.01 0.00 0.01 -0.87 3.74WY05 -0.02 -0.02 0.02 0.32 1.73WY06 0.00 0.00 0.01 -0.26 7.36WY07 0.03 0.02 0.01 0.42 2.83WY08 -0.01 0.00 0.01 -0.33 7.33WY09 0.01 0.01 0.01 0.35 6.36WY10 -0.02 0.00 0.01 -1.60 5.85WY11 0.02 0.01 0.01 0.90 5.95Average 0.00 0.00 0.01 -0.13 5.144.4.2 Erosion and Deposition HistogramsBed material erosion and deposition were analyzed separately for the riffle andpool units in East Creek in the rapids reach and in all riffle pool sub-reaches forWY04-10.RifflesErosion histograms for the riffles were generally uni-modal with a few cases ofmulti-modal histograms in each sub-reach. These consisted of three multi-modalhistograms in the rapids in WY07, WY08, and WY10; three multi-modal his-tograms in riffle pool 1 in WY07, WY09, and WY10; three multi-modal histogramsin riffle pool 2 in WY04, WY07, and WY10; and one multi-modal histogram inriffle pool 3 in WY05. Deposition histograms for the riffles were generally uni-modal; however, there was a higher occurrence of multi-modal histograms of depo-sition compared with erosion. These multi-modal deposition histograms consistedof four cases in the rapids in WY04, WY08, WY09, and WY10; one case in rifflepool 1 in WY04; four cases in riffle pool 2 in WY04, WY05, WY07, and WY11;and three cases in riffle pool 3 in WY04, WY06, and WY07.42Figure 4.10 shows the erosion and deposition histograms in riffle pool 3 inWY09 as an example of a uni-modal erosion histogram concurrent with a uni-modal deposition histogram. Figure 4.11 shows the erosion and deposition his-tograms in riffle pool 1 in WY07 as an example of a bimodal erosion histogramconcurrent with a uni-modal deposition histogram Figure 4.12 shows the erosionand deposition histograms in the rapids in WY10 as an example of a multimodalerosion histogram concurrent with a multi-modal deposition histogram. Multi-modal histograms occurred most commonly in WY07 and WY10 for erosion andin WY04 for deposition. Occurrence of a multi-modal erosion histogram was notconsistent with the occurrence of a multi-modal deposition histogram for the samegiven year and sub-reach.Erosion and deposition histograms for all reaches and all years can be found inAppendix B in Figures B.1, B.2, B.3, and B.4 for the rapids, riffle pool 1, rifflepool 2, and riffle pool 3, respectively.Figure 4.10: Uni-modal erosion and deposition histograms in riffle pool 3 forriffles in WY0943Figure 4.11: A bi-modal erosion histogram concurrent with a uni-modal de-position histogram in riffle pool 1 for riffles in WY07Figure 4.12: Multimodal erosion and deposition histograms in the rapids forriffles in WY10The average mean erosion (-0.04 m) was equal in absolute magnitude to theaverage mean deposition (0.04 m) in riffles. Similarly, the average median erosion(-0.03 m) was equal in absolute magnitude to the average median deposition (0.03m) in riffles. The mean erosion ranged from -0.12 m to -0.02 m and the meandeposition ranged from 0.02 m to 0.08 m. The median erosion ranged from -0.06m to -0.02 m and the median deposition ranged from 0.01 m to 0.07 m.44The average variances of both the erosion and deposition histograms were0.003 m2. The largest variances in the erosion histograms occurred during WY05for three of four sub-reaches and the largest variances in the deposition histogramsoccurred in WY07 for three of four sub-reaches. The channel experienced largestorm events in WY07. The variance for erosion ranged from 0.000 m2 to 0.029m2 and the variance for deposition ranged from 0.000 m2 to 0.012 m2.All erosion histograms were negatively skewed and all deposition histogramswere positively skewed for the riffles. The average skew of the erosion histogramswas -2.84 and the average skew of the deposition histograms was 2.96. The largestskews for the erosion histograms occurred in WY08 in the rapids, riffle pool 2, andriffle pool 3. The smallest skews for the erosion histograms occurred in WY07 inthe riffle pool 1, riffle pool 2, and riffle pool 3 sub-reaches. Unlike the erosion his-tograms, the most extreme skews for the deposition histograms did not repeatedlyoccur in the same water year across the sub-reaches. Rather, the smallest skewsand the largest skews for the deposition histograms occurred in different wateryears across the sub-reaches.There was a wide range in histogram kurtosis, with an average kurtosis of 13.9for erosion histograms and an average kurtosis of 13.8 for deposition histograms.The minimum (-0.70) and maximum (39.36) kurtosis values for erosion both oc-curred in riffle pool 2. The minimum (2.12) and maximum (46.25) kurtosis valuesfor deposition occurred in riffle pool 2 and riffle pool 1, respectively.The summary statistics for erosion in riffles in riffle pool 2 are shown as arepresentative for other sub-reaches (Table 4.7). Summary statistics for erosionfor the remaining reaches can be found in Appendix B in Tables B.1, B.2, B.3for the rapids, riffle pool 1, and riffle pool 3, respectively.45Table 4.7: Summary statistics for erosion in riffles in riffle pool 2 sub-reachYear Mean Median Variance Skew Kurtosis(m) (m) (m2)WY04 -0.04 -0.03 0.00 -1.22 1.14WY05 -0.06 -0.06 0.00 -1.63 4.94WY06 -0.02 -0.02 0.00 -4.53 33.36WY07 -0.07 -0.06 0.00 -0.46 -0.70WY08 -0.03 -0.02 0.00 -5.51 39.36WY09 -0.03 -0.02 0.00 -3.60 26.73WY10 -0.04 -0.02 0.00 -2.49 8.30WY11 -0.03 -0.02 0.00 -2.04 8.07Average -0.04 -0.03 0.00 -2.69 15.15The summary statistics for deposition in riffles in riffle pool 2 are shown as arepresentative for the other sub-reaches (Table 4.8). Summary statistics for depo-sition for the remaining reaches can be found in Appendix B in Tables B.4, B.5,B.6 for the rapids, riffle pool 1, and riffle pool 3, respectively.Table 4.8: Summary statistics for deposition in riffles in riffle pool 2 sub-reachYear Mean Median Variance Skew Kurtosis(m) (m) (m2)WY04 0.04 0.02 0.00 2.74 8.32WY05 0.04 0.03 0.00 1.34 4.38WY06 0.02 0.02 0.00 2.95 15.08WY07 0.08 0.07 0.00 1.37 2.12WY08 0.02 0.01 0.00 3.73 22.63WY09 0.03 0.02 0.00 3.19 15.76WY10 0.02 0.02 0.00 3.25 20.75WY11 0.03 0.02 0.00 3.30 15.43Average 0.04 0.03 0.00 2.73 13.0646PoolsErosion histograms for the pools were generally uni-modal with a few cases ofmulti-modal histograms in each sub-reach. These consisted of four multi-modalhistograms in the rapids in WY04, WY06, WY08, and WY10; two multi-modalhistograms in riffle pool 1 in WY07 and WY08; three multi-modal histograms inriffle pool 2 in WY05, WY06, and WY08; and two multi-modal histograms in rifflepool 3 in WY04 and WY11. Deposition histograms for the pools were generallyuni-modal with a few cases of multi-modal histograms in each sub-reach. Thesemulti-modal deposition histograms consisted of four cases in the rapids in WY04,WY06, WY07, and WY10; one case in riffle pool 1 in WY11; two cases in rifflepool 2 in WY07 and WY09; and one case in riffle pool 3 in WY05.Figure 4.13 shows the erosion and deposition histograms in riffle pool 3 inWY09 as an example of a uni-modal erosion histogram concurrent with a uni-modal deposition histogram. Figure 4.14 shows the erosion and deposition his-tograms in the rapids in WY06 as an example of a multi-modal erosion histogramconcurrent with a bi-modal deposition histogram. Figure 4.15 shows the erosionand deposition histograms in riffle pool 2 in WY05 as an example of a multimodalerosion histogram concurrent with a uni-modal deposition histogram. Occurrenceof multi-modal erosion and deposition histograms was fairly spread out across theyears of study, with W09 being the only water year in which all histograms wereuni-modal. Occurrence of a multi-modal erosion histogram was not consistent withthe occurrence of a multi-modal deposition histogram for the same given year andsub-reach.Erosion and deposition change histograms for all reaches and all years can befound in Appendix B in Figures B.1, B.2, B.3, and B.4 for the rapids, riffle pool1, riffle pool 2, and riffle pool 3, respectively.47Figure 4.13: Uni-modal erosion and deposition histograms in riffle pool 3 forpools in WY09Figure 4.14: A multi-modal erosion histogram concurrent with a bi-modaldeposition histogram in the rapids for pools in WY0648Figure 4.15: Multi-modal erosion and uni-modal deposition histograms inriffle pool 2 for pools in WY05The average mean erosion (-0.06 m) was equal in absolute magnitude to theaverage mean deposition (0.06 m) in riffles. Similarly, the average median erosion(-0.04 m) was equal in absolute magnitude to the average median deposition (0.04m) in riffles. The mean erosion ranged from -0.12 m to -0.03 m and the meandeposition ranged from 0.03 m to 0.14 m. The median erosion ranged from -0.09m to -0.02 m and the median deposition ranged from 0.02 m to 0.12 m.The average variance of erosion and deposition histograms were 0.005 m2 and0.004 m2, respectively. The largest variances in erosion and deposition histogramswere not consistent with water year. The variance for erosion ranged from 0.001m2 to 0.012 m2 and the variance for deposition ranged from 0.001 m2 to 0.011 m2.All erosion histograms were negatively skewed and all deposition histogramswere positively skewed for the riffles. The average skew of the erosion histogramswas -2.36 and the average skew of the deposition histograms was 2.46. The mostextreme skews for both erosion and deposition histograms for pools occurred indifferent water years across the sub-reaches.There was a wide range in histogram kurtosis, with average kurtosis values of8.2 and 9.0 for erosion and deposition histograms, respectively. The minimum kur-tosis for erosion (0.74) occurred in the rapids in WY07 and the maximum kurtosisfor erosion (19.06) occurred in riffle pool 1 in WY11. The minimum (0.71) andmaximum (24.67) kurtosis values for deposition both occurred in riffle pool 2 in49WY07 and WY11, respectively.The summary statistics for erosion in pools in riffle pool 2 are shown as arepresentative for other sub-reaches (Table 4.9). Summary statistics for erosionfor the remaining reaches can be found in Appendix B in Tables B.7, B.8, B.9for the rapids, riffle pool 1, and riffle pool 3, respectively.Table 4.9: Summary statistics for erosion in pools in riffle pool 2 sub-reachYear Mean Median Variance Skew Kurtosis(m) (m) (m2)WY04 -0.07 -0.04 0.01 -3.44 16.13WY05 -0.12 -0.09 0.01 -1.20 1.19WY06 -0.06 -0.03 0.01 -2.44 7.87WY07 -0.10 -0.08 0.01 -1.67 3.91WY08 -0.06 -0.03 0.00 -2.56 10.60WY09 -0.05 -0.03 0.00 -2.58 9.17WY10 -0.05 -0.02 0.01 -3.02 11.32WY11 -0.05 -0.03 0.00 -1.84 4.49Average -0.07 -0.04 0.01 -2.34 8.09The summary statistics for deposition in pools in riffle pool 2 are shown asa representative for the other sub-reaches (Table 4.10). Summary statistics fordeposition for the remaining reaches can be found in Appendix B in Tables B.10,B.11, B.12 for the rapids, riffle pool 1, and riffle pool 3, respectively.4.4.3 Bed Storage, Erosion, and DepositionBed material erosion and deposition were compared to bed material storage sepa-rately for the riffle and pool units in East Creek for WY04-11 both distinctly foreach of the upper reaches and in the upper reaches combined. It was hypothe-sized that storage should track trends in the erosion/deposition balance for eachsub-reach.Figure 4.16 shows weak correlations between net change in bed elevation andmean bed erosion and deposition in riffles. Similarly, Figure 4.17 shows weak cor-relations between net change in bed elevation and mean bed erosion and depositionin pools.50Table 4.10: Summary statistics for deposition in pools in riffle pool 2 sub-reachYear Mean Median Variance Skew Kurtosis(m) (m) (m2)WY04 0.07 0.05 0.01 2.03 5.13WY05 0.07 0.05 0.00 2.08 5.76WY06 0.05 0.03 0.00 2.90 10.47WY07 0.14 0.12 0.01 0.95 0.71WY08 0.05 0.03 0.00 2.10 5.90WY09 0.07 0.04 0.01 2.93 12.43WY10 0.05 0.03 0.00 2.12 6.60WY11 0.05 0.03 0.01 4.13 24.67Average 0.07 0.05 0.01 2.41 8.96y = 0.6379x -  0.0436 R² = 0.184  y = 0.5426x + 0.0427  R² = 0.1723  -0.15-0.1-0.0500.050.1-0.0500 -0.0400 -0.0300 -0.0200 -0.0100 0.0000 0.0100 0.0200 0.0300 0.0400Mean Bed Erosion and Deposition (m) Net Change in Bed Elevation (m)  ErosionDepositionFigure 4.16: Sediment storage vs. erosion and deposition in riffles in com-bined upper reaches of East Creek51y = 0.4698x -  0.0594 R² = 0.2188  y = 0.2623x + 0.058  R² = 0.0677  -0.15-0.1-0.0500.050.10.150.2-0.08 -0.06 -0.04 -0.02 0 0.02 0.04 0.06 0.08Mean Bed Erosion and Deposition (m) Net Change in Bed Elevation (m)  ErosionDepositionFigure 4.17: Sediment storage vs. erosion and deposition in pools in com-bined upper reaches of East Creek4.4.4 Peak Discharge, Erosion, and DepositionBed material erosion and deposition were compared to annual peak discharge sep-arately for the riffle and pool units in the rapids reach and riffle pool 1, 2, and 3sub-reaches of East Creek for WY04-11.In most sub-reaches as peak discharge increased, the magnitude of erosion anddeposition increased. Figure 4.18 shows the relationship between peak dischargeand mean bed erosion and the relationship between peak discharge and mean beddeposition for riffles in the riffle pool 3 sub-reach. This location is shown becausethe slopes and R2 values are fairly representative of those observed in the otherunits and sub-reaches analyzed. Figure 4.19 shows an exception to this trend in thepools in the rapids reach. In this case, as peak discharge increased, the magnitudeof erosion decreased slightly.52y = - 0.0013x -  0.0267 R² = 0.0473  y = 0.005x + 0.0178  R² = 0.2243  -0.06-0.04-0.0200.020.040.060.080.000 0.500 1.000 1.500 2.000 2.500 3.000 3.500 4.000 4.500 5.000Mean Bed Erosion and Deposition (m) Peak Discharge (m 3/s) ErosionDepositionFigure 4.18: Peak discharge vs. mean bed erosion and mean bed depositionin riffles of riffle pool 3 sub-reach in East Creek53y = 0.0012x -  0.0443 R² = 0.0145  y = 0.0041x + 0.0305  R² = 0.3466  -0.08-0.06-0.04-0.0200.020.040.060.080.000 1.000 2.000 3.000 4.000 5.000Mean Bed Erosion and Deposition (m) Peak Discharge (m 3/s) ErosionDepositionFigure 4.19: Peak discharge vs. mean bed erosion and mean bed depositionin pools of rapids reach in East CreekThe correlations between peak discharge and mean bed deposition (Table 4.11)were higher than the correlations between peak discharge and mean bed erosion(Table 4.12) in all sub-reaches. The largest correlations for deposition and forerosion occurred in riffle pool 2 as highlighted by the bolded font.Table 4.11: Correlation between peak discharge and mean bed depositionMorphological Unit Rapids Riffle Pool 1 Riffle Pool 2 Riffle Pool 3Pools 0.3466 0.1864 0.4278 0.3692Riffles 0.1883 0.3924 0.4287 0.224354Table 4.12: Correlation between peak discharge and mean bed erosionMorphological Unit Rapids Riffle Pool 1 Riffle Pool 2 Riffle Pool 3Pools 0.0145 0.0358 0.0602 0.002Riffles 0.0242 0.0659 0.1832 0.0473The correlations between peak discharge and median bed erosion and the cor-relations between peak discharge and median bed deposition were also low (R2 ≤0.5451) in all sub-reaches of East Creek.55Chapter 5DiscussionThe discussion section proposes possible explanations and offers comments on thesediment storage and transport trends presented in Chapter 4: Results. First, flowregime is briefly discussed. Next, sediment storage and transport processes in EastCreek at the channel, reach, and unit scales are discussed. At the channel scale,bed elevation changes in East Creek throughout the study period are described. Atthe reach scale, emphasis is given to analyzing sediment storage trends over timeand space. At the unit scale, the ability of the data to demonstrate conservation ofmass in East Creek is discussed.5.1 Flow RegimeIn considering the relationship between discharge and sediment transport in thesections that follow, it is important to acknowledge the scale and scope at whichthe flow regime was analyzed in this study. Consistent with the annual scale ofthe sediment budget, discharge was also analyzed at the annual scale where annualpeak discharge was used as a coarse reflection of annual flow regime. Since dis-charge is not resolved at the event scale, the impact of event scale features of flowregime are not considered. For example, number of storm events per year; stormduration; and sediment mobilizing flow duration per storm event are not addressed.Despite this limitation, annual peak discharge does provide valuable informationabout the sediment mobilizing conditions in East Creek. In particular, the 200756and 2009 water years experienced the highest annual peak discharges (4.3 m3/sand 4.5 m3/s, respectively), suggesting high sediment mobilizing capacity in theseyears. In comparison, the 2006 water year experienced the lowest peak discharge(1.0 m3/s) (Figure 4.1). Overall, the peak flows fluctuated over time between highand low values.5.2 Channel Scale: Long ProfileBed elevation in the upper reaches fluctuated between 2003 and 2009. At thecoarser scale, the long profile of East Creek suggested relative channel stability(Figure 4.2); whereas, at the finer scale additional bed elevation changes couldbe discerned (Figure 4.3). In the rapids reach, the greatest amount of fluctuationsoccurred in the upper 20 m of the reach (Figure 4.3a). The channel was scoureddown to the bedrock in the upper 20 m during WY07, the year with the great-est storm events. Downstream sediment accumulation in subsequent years in rifflepools 2 and 3 can likely be attributed to the re-mobilization of the sediment scouredfrom the upper 20 m of the rapids (Figures 4.3c and 4.3d). Fluctuations in bedelevation were of similar magnitude across the sub-reaches. There were, however,localized areas distributed throughout the upper reaches in which more substantialfluctuations in bed elevation occurred. These large fluctuations occurred acrossvarious bedforms: runs, pools, and bars, and across varying channel widths rang-ing from 1.9 m to 6.2 m. Large bed elevation fluctuations frequently occurred inclose proximity to a channel obstruction downstream of the culvert and associatedplunge pool at the top of the rapids section and around large wood in the rifflepool sub-reaches. It is well established that wood modulates sediment flux andstorage through a coupled cycle of storage-erosion and release (Eaton et al., 2012).Sediment is stored upstream of wood, the presence of wood leads to bed erosionalongside and downstream of the wood, and when wood decays stored sediment isreleased to downstream reaches.575.3 Reach Scale: Closing the Sediment Budget5.3.1 Bed Load Flux and StorageThere was considerable variation in the magnitude of bed load flux, with no con-sistent connections to sub-reach or water year, suggesting that channel morphologyand water discharge are not the only factors controlling bed load flux in East Creek.The TOR, the sub-reach just upstream of the rapids sub-reach (Figure 5.1),experienced the greatest annual bed load flux, which can likely be attributed to thepresence of a culvert at its upper bound.Figure 5.1: Upper reaches of East Creek with location of top of rapids (TOR)sub-reach highlighted in red circleThe narrow diameter of the culvert in comparison to the channel width resultsin a localized increase in water velocity, and consequently an increase in sedimentmobilizing capacity in this short stretch of the channel. This localized increase invelocity is also evidenced by the presence of a large scoured out plunge pool atthe base of the culvert (Figure 5.2). It is likely that some of the material scouredfrom the plunge pool contributed to the material captured in the trap a mere 11.9 mdownstream.58Photo Credit: Joshua CaulkinsFigure 5.2: Plunge pool at base of culvert in top of rapids (TOR) sub-reachof East CreekBoth the largest average annual bed load flux and the largest bed load stor-age over the 3 year period occurred in the rapids reach. This dynamism withinthe rapids reach is consistent with the earlier observation that the rapids reach un-derwent the largest amount of change in the thalweg elevation over the nine yearperiod (Section 5.2).The majority of sub-reaches experienced positive net bed load storage suggest-ing that during WY09-11 East Creek may have been in a state of aggradation 4.3.The exceptions to this occurred in the rapids in WY11 (-651 kg) and in riffle pool592 in WY10 (-849 kg). The uncharacteristic negative storage observed in the rapidsin WY11 may have been a response to the particularly large positive sediment stor-age (1252 kg) in WY10. That is, to balance out the large gain in sediment in theprevious year, more sediment than usual was excavated from the reach in the fol-lowing year. With regards to the uncharacteristic negative storage observed in rifflepool 2 in WY10, it is possible that this is a reflection of a reduction in sedimentsupply to this sub-reach. Since riffle pool 2 is downstream of riffle pool 1 and therapids, sediment excavated from these more upstream reaches is often transporteddownstream and deposited in riffle pool 3. In the 2010 water year, there were par-ticularly low peak flows; and, correspondingly, particularly high bed load storagevalues in the rapids and riffle pool 1. With this increase in sediment retention in theupper reaches, less sediment would have been available to travel downstream andbe deposited in the riffle pool 3 sub-reach, resulting in negative bed load storage.5.3.2 Bed Material Erosion and StorageSimilar to bed load flux and storage, there was considerable variation in the mag-nitude of bed material erosion and storage, with no consistent connection to sub-reach or water year, suggesting that channel morphology and water discharge arenot the only factors controlling bed material erosion and storage in East Creek. Itis likely that a third, and important, control on bed material erosion and storage inEast Creek is sediment supply and storage within the channel. Sediment scouredfrom upstream reaches becomes a sediment source in downstream reaches, andsediment storage fluctuations in a given sub-reach and year can be more easily ex-plained by examining bed conditioning. Given that a longer record of bed materialerosion and storage is available compared to bed load flux and storage, a moredetailed analysis of bed material storage changes over time and space is possible.Over the eight year period analyzed, all sub-reaches fluctuated between aggra-dational and degradational states (Figures 4.5a - 4.5d). The rapids were dominatedby degradation during the early years and transitioned towards being slightly dom-inated by aggradation in the latter half of the study (Figure 4.5a). Consistent withthe rapids being the reach where the largest bed load flux and storage occurred, therapids was also the reach where the largest average bed material flux and storage60occurred. Given that (1) the thalweg elevation in the rapids was relatively stablein the lower part of the reach, but fluctuated substantially in the upper part of thereach (Figure 4.3a) and (2) in the upper part of the reach, there was scouring tosuch an extent that the bedrock was exposed (Figure 5.3), it is likely that the largebed material flux values in the rapids, too, were localized in the upper part of thereach. The bed material scoured from the upper part of the reach was likely trans-ported and deposited in the lower part of the reach contributing to the large bedmaterial storage values in the rapids. Similar to the storage patterns over time inthe rapids, riffle pool 1 was also dominated by degradation in the early years andtransitioned towards being slightly dominated by aggradation in the latter half ofthe study (Figure 4.5b). Like the rapids, the greatest change in storage (nega-tive storage in both cases) occurred in WY07, corresponding to the second highestannual peak discharge.Photo Credit: Joshua CaulkinsFigure 5.3: Exposure of till on bed of rapids reach of East Creek in WY07following high magnitude scouringIn contrast to the trend from degradation to aggradation over time observed inthe rapids and riffle pool 1, riffle pools 2 and 3 fluctuated back and forth betweenaggradational and degradational states throughout the entire duration of the study61period. The magnitudes of annual storage in riffle pool 2 spanned a narrower rangethan that of the other sub-reaches, with no water year experiencing a drasticallyhigher storage than the others (Figure 4.5c). Similar to riffle pool 2, in riffle pool3, storage values were closer in magnitude to each other over time compared to thatof the rapids and riffle pool 1. The greatest storage in riffle pool 3 (positive storage)occurred in WY11, a year without a particularly high annual peak discharge (Figure4.5d).Analyzing the spatial trends across the length of the channel for each water yearmay offer more insight into some of the temporal trends noted above (Figures 4.6- 4.7). Moving downstream along the length of the channel, there was increasingdegradation approaching riffle pool 3 in WY04. This was followed by fluctuatingstorage values in WY04 along the length of the channel. In the following year,WY05 there were also storage fluctuations along the length of the channel, but inthe opposite directions as the previous year, suggesting a possible internal balanc-ing of storage occurring in the study reach. For example, in WY05 there werepeaks in storage in RP1 and RP3 and troughs in the RAP and RP2, whereas inWY05 there were peaks in storage in the RAP and RP2 and troughs in RP1 andRP3. In WY07, the year with the second highest annual peak discharge, therewas an exceptionally large amount of scour in the rapids and moving downstreamalong the length of the channel increasing aggradation, suggesting that some of thematerial scoured out from the rapids may have been transported downstream anddeposited in riffle pools 2 and 3. This increasing trend towards aggradation movingdownstream along the channel was then balanced out the following year, in WY08,with a trend towards increasing degradation moving downstream along the lengthof the channel. It is presumable that the degradation in riffle pools 2 and 3 in WY08may be a consequence of the large input of sediment in the previous year. The largeload of sediment input to the downstream reaches from the rapids in WY07 waslikely then excavated in WY08, producing negative sediment storage values. InWY09, the entire channel shifted to a state of aggradation, with a particularly largeincrease in storage in the more downstream sub-reaches. This is surprising giventhat the greatest annual peak flow occurred in WY09 and suggests that (1) annualpeak flow cannot be used as a sole predictor of sediment storage and/or (2) theannual peak flow values used in this study may be suspect. In WY09, the storage62values returned to fluctuating between aggradation and degradational states mov-ing downstream along the length of the channel implying in-stream balancing outof sediment storage. Finally, in WY11 the channel was in a state of aggradationwith increasing positive storage values moving downstream along the length of thechannel.5.3.3 Comparison of MethodsThere was high discrepancy between sediment storage estimates obtained fromtraps compared to those obtained from surveyed difference maps, with the percent-age differences between these two methods ranging from -88% to 6030% (Table5.1). With the exception of the rapids in WY11 (percentage difference of -88%), inall other sub-reaches and water years analyzed the pit trap method yielded smallermagnitudes of sediment storage than the difference mapping method. Since the pittrap method measures bed load and the difference mapping method measures bedmaterial load, this suggests that there was more bed material load compared to bedload. This can be explained in at least two ways. First, when the traps becomefull or almost full finer material that is coupled to the flow overpasses the traps anddoes not contribute to the pit trap sediment collections. That same fine materiallikely would have settled out elsewhere in the channel, perhaps in bars or riffles,contributing to the higher values of bed material load calculated. Second, the bedload traps have storage volumes that reflect a fraction of the bed material that is ac-tually transported during any given storm or runoff period. Hence, the trap recordswill always underestimate bed material flux.63Table 5.1: Sediment budget from traps and survey data for East Creek for W09-11Station Water Year Sediment Storage Sediment Storage Difference % Differencefrom Traps from Surveys between between(I-E) (∆S) (I-E) and ∆S (I-E) and (∆S)(kg) (kg) (kg) (%)RAP WY09 156 1554 1398 896RAP WY10 1252 1641 389 31RAP WY11 -651 -80 572 -88RP1 WY09 264 900 636 241RP1 WY10 893 2763 1870 209RP1 WY11 532 1006 474 89RP2 WY09 29 1778 1749 6030RP2 WY10 -849 -5497 -4648 548RP2 WY11 147 3233 3086 210564Figure 5.4 shows that in all but two cases, the sediment storage estimates ob-tained from the morphological mapping method are equal to or greater than doublethose obtained from the pit trap method. This highlights the extent to which thepit trap method comparatively underestimates sediment storage, regardless of yearor morphological reach. This is consistent with the earlier supposition of pit trapinefficiency.-6000-5000-4000-3000-2000-100001000200030004000-6000.0 -4000.0 -2000.0 0.0 2000.0 4000.0Sediment storage estimates  from surveys  (ΔS) (kg) Sediment storage estimates from traps (I - E ) (kg)  RA PRP1RP21:1 line2:1 lineFigure 5.4: Sediment storage estimates from surveys compared to sedimentstorage estimates from trapsA closer look at the two exceptions to this trend of the pit trap method greatlyunderestimating storage compared to the morphological mapping method may of-fer more insight or create further questions about pit trap efficiency. The first casein which the sediment storage estimate obtained from the morphological mappingmethod was less than double that obtained from the pit trap method occurred inthe rapids in WY10. This coincided with the highest recorded sediment storage65estimates as obtained using the pit trap method of all sub-reaches and water yearswithin the given study period, explaining why this estimate may have more closelyapproached that of the morphological mapping method. In WY10 there were doc-umented trap overflow events in the rapids making for an incomplete accounting ofsediment storage using the pit traps. It is impossible to draw absolute conclusionsfrom a single case, but this may suggest that limitations in trap capacity contributeto underestimates of sediment storage, but cannot fully explain the more than two-fold differences observed between the two methods. The second exception to thepit trap underestimates trend occurred in riffle pool 2 in WY10. This was the onlycase in which the sediment storage estimates from the pit trap were greater thanthose obtained from the morphological mapping method. This coincided with thelowest recorded sediment storage estimate obtained from the morphological map-ping method of all sub-reaches and water years analyzed and represented one ofonly two cases of negative sediment storage estimates. Framed another way - thatthe morphological mapping method captured the erosion of a larger fraction of bedmaterial than the pit trap could be capable of doing - it becomes less surprising thatthe pit trap estimate of storage was higher than that of the morphological mappingmethod in this case.Figure 5.5 shows that the difference in sediment storage estimates between themorphological mapping and pit trap method can be scaled to the sediment storageestimates obtained from the morphological mapping method with a correlation of0.95. This suggests that it might be possible to somewhat standardize the magni-tude of discrepancy between these two methods. This could be very valuable for(1) increasing the comprehensiveness of a sediment budget and (2) more accuratelyestimating the magnitude of over- or under- estimation of sediment storage whenonly one of these two methods is available.66y = 0.822x -  52.554 R² = 0.9492  -6000.0-5000.0-4000.0-3000.0-2000.0-1000.00.01000.02000.03000.04000.0-6000 -4000 -2000 0 2000 4000Difference between   sediment storage estimates from traps (I-E) and  sediment storage estimates from surveys (ΔS)  (kg) Sediment storage estimates from surveys ( Δ S)   (kg)  Figure 5.5: Difference in sediment storage estimates between two distinctmethods (surveys and traps) scaled to sediment storage estimates fromsurveys5.4 Unit Scale: Sediment Storage and Bedforms5.4.1 Bed Elevation Change in Riffles and PoolsIn most cases, the bed elevation change histograms were relatively symmetricalwith means and medians falling close to 0.0 m. This suggests that erosion anddeposition were balanced, demonstrating mass conservation at the reach scale andmore importantly that erosion and deposition are coupled. The level of detail pre-sented in the histograms provides a unique opportunity to observe conservation ofmass at the unit scale over an extended time period that is typically not seen instudies on sediment transport.The level of detail presented in the data also allows for spatial patterns of bed67elevation change across riffles and pools to be discerned. The means and mediansof bed elevation change were close to 0.0 m in most cases in both the riffle and poolunits. There were positive linear correlations between mean bed elevation changein riffles and mean bed elevation change in pools in all sub-reaches except rifflepool 1. This is likely a reflection of the discharge and sediment supply conditionsin the channel. There were only two years of the study period during which partic-ularly high annual peak flows capable of causing extreme scouring were observed.Consequently, during the majority of the study period, mean scour and fill wererelatively balanced across the channel, producing similarly low mean bed eleva-tion change in both riffles and pools. There were, however, weak or no correlationsbetween median bed elevation change in riffles and median bed elevation change inpools in all reaches (Table 5.2), suggesting that finer scale bed elevation changesvary across riffles and pools.Table 5.2: Correlation between bed elevation change in riffles and bed eleva-tion change in pools in upper reaches of East CreekMeasure of Bed Rapids Riffle Pool 1 Riffle Pool 2 Riffle Pool 3Elevation ChangeMean 0.82 0.05 0.81 0.70Median 0.38 0.00 0.90 0.09In the rapids reach, variances in bed elevation change were higher in rifflesthan in pools. In contrast, in all riffle pool sub-reaches, variances in bed eleva-tion change were greater in pools than in riffles (Figure 5.6). There may havebeen increased variance in bed elevation change in pools compared to riffles in theriffle pool morphology because pools characteristically contain finer more looselyinteracting particles than riffles, allowing for increased sediment mobility and con-sequently, greater variance in bed elevation changes. The increased variance inbed elevation change in riffles compared to pools in the rapids morphology is moredifficult to explain from the perspective of particle interactions; however, it can beexplained by broadening consideration to include reach scale processes. The over-whelming majority of bed elevation changes in the rapids were localized in the68upper 20 m of the reach, which primarily encompassed a riffle unit. This concen-tration of high magnitude, high variance, bed elevation change localized in a riffleunit likely masked smaller scale spatial variations in bed elevation change acrossriffle and pool units distributed throughout the reach.6900.0050.010.0150.020.02504 05 06 07 08 09 10 11Variance Water Year (a) Rapids00.0020.0040.0060.0080.010.0120.0140.0160.01804 05 06 07 08 09 10 11Variance Water Year (b) Riffle Pool 100.0050.010.0150.020.02504 05 06 07 08 09 10 11Variance Water Year (c) Riffle Pool 200.0020.0040.0060.0080.010.0120.0140.0160.01804 05 06 07 08 09 10 11Variance Water Year (d) Riffle Pool 3Figure 5.6: Variance in bed elevation over time in riffles (purple line) and pools (blue line) in the rapids, riffle pool 1,riffle pool 2, and riffle pool 3 sub-reaches of East Creek70The direction of skew in both riffles and pools fluctuated over time, as wouldbe expected for a channel in a relative state of equilibrium. In some cases thedirection of skew of the pools very crudely appeared to oscillate in phase with thedirection of skew in the riffles; however, this pattern was inconsistent and therewere very weak or no correlations between skew in riffles and skew in pools forall sub-reaches (R2 ≤ 0.2). Figure 5.7 shows an example of the bed elevationhistogram skew in riffles and pools in riffle pool 1, a case in which the crudely in-phase oscillation can be observed. Figure A.5 in Appendix B shows the skew ofthe bed elevation histograms over time for all sub-reaches. There were very weak(R2 = 0.02) or no correlations between kurtosis in riffles and kurtosis in pools forall sub-reaches.5.4.2 Erosion and Deposition in Riffles and PoolsWith the exception of the riffle pool 2 sub-reach, there was very little correlationbetween annual deposition in riffles and annual deposition in pools. Similarly,with the exception of the riffle pool 2 sub-reach, there was also very little corre-lation between annual erosion in riffles and annual erosion in pools (Table 5.3).This observation strongly reinforces the importance of considering spatial hetero-geneity in sediment transport at the unit scale. Even when a morphological reachwas subjected to the same flow regime, bed conditioning, and sediment supplyconditions, there were still localized differences in erosion and deposition patternswithin the reach. These differences were likely in part governed by differing graininteractions in riffles and pools.Table 5.3: Correlation between erosion/deposition in riffles and erosion/de-position in pools in upper reaches of East CreekType of Change Rapids Riffle Pool 1 Riffle Pool 2 Riffle Pool 3Deposition 0.4178 0.7789 0.9559 0.3328Erosion 0.5755 0.2937 0.8512 0.5481In the riffle pool sub-reaches, the magnitude of deposition in pools was al-ways higher than deposition in riffles. Whereas, in the rapids reach, the magnitude71-2-1.5-1-0.500.511.522.5304 05 06 07 08 09 10 11Skewness Water Year Figure 5.7: Skew of bed elevation histograms over time in riffles (purple line)and pools (blue line) in riffle pool 1 in East Creekof deposition in pools was always lower than deposition in riffles. The erosionshowed a similar pattern. In the riffle pool sub-reaches, the magnitude of erosionin pools was nearly always higher than erosion in riffles. Whereas, in the rapidsreach, the magnitude of erosion in pools was always lower than erosion in rif-fles. These observations build on the earlier observation in Subsection 5.4.1 ofincreased variance in bed elevation change in pools compared to riffles in the rifflepool morphology. As with variance, greater magnitudes of deposition and erosionin pools compared to riffles in the riffle pool morphology are likely a reflection ofcomparatively increased sediment mobility in pools, where particles are finer andmore loosely interacting than in riffles. As with variance, the opposite trend wasobserved for the rapids, where there was a smaller magnitude of deposition anderosion in pools compared to riffles. In the rapids morphology, reach scale pro-72cesses were likely a greater determinant of spatial patterns of sediment transport;whereas, in the riffle pool morphology, unit scale processes were likely a greaterdeterminant of spatial patterns of sediment transport. This is a wonderfully clearillustration of the necessity of integrating multiple spatial scales when predictingsediment transport patterns.In riffles and pools in all upper reaches, deposition and erosion fluctuated in-consistently with respect to water year (Figure 5.8). In the majority of cases,the greatest magnitudes of erosion and deposition in riffles and pools occurred inWY07, the year of the second highest annual peak flows. There did not appear tobe any other connections between water year and magnitude of deposition/erosionin riffle or pool units, a reinforcement that flow regime cannot alone be used as areliable predictor of sediment transport patterns.7300.020.040.060.080.104 05 06 07 08 09 10 11Mean Bed Deposition (m) Water Year (a) Deposition: RAP00.020.040.060.080.104 05 06 07 08 09 10 11Mean Bed Deposition (m) Water Year (b) Deposition: RP100.050.10.150.204 05 06 07 08 09 10 11Mean Bed Deposition (m) Water Year (c) Deposition: RP200.020.040.060.080.104 05 06 07 08 09 10 11Mean Bed Deposition (m) Water Year (d) Deposition: RP3-0.14-0.12-0.1-0.08-0.06-0.04-0.020 04 05 06 07 08 09 10 11Mean Bed Erosion (m) Water Year (e) Erosion: RAP-0.12-0.1-0.08-0.06-0.04-0.020 04 05 06 07 08 09 10 11Mean Bed Erosion (m) Water Year (f) Erosion: RP1-0.14-0.12-0.1-0.08-0.06-0.04-0.020 04 05 06 07 08 09 10 11Mean Bed Erosion (m) Water Year (g) Erosion: RP2-0.1-0.08-0.06-0.04-0.020 04 05 06 07 08 09 10 11Mean Bed Erosion (m) Water Year (h) Erosion: RP3Figure 5.8: Mean bed deposition and mean bed erosion over time in the riffle units (purple line) and the pool units(blue line) in the rapids, riffle pool 1, riffle pool 2, and riffle pool 3 sub-reaches of East Creek745.4.3 Bed Storage, Erosion, and DepositionAlthough it was hypothesized that storage should track trends in erosion/deposi-tion in riffles and pools, there were very weak correlations between net change inbed elevation and mean bed erosion/deposition. Taking riffles as an example, inyears during which there was a large amount of storage in riffles, there was notnecessarily an associated increase in mean deposition (Figure 4.16). For storageto increase independently of deposition, there must have been a decrease in ero-sion. Similarly, in years during which there was very little storage in riffles, therewas not necessarily an associated increase in mean erosion. For storage to decreaseindependently of erosion, there must have been a decrease in deposition. A similartrend was observed for pools (Figure 4.17). This implies that erosion and deposi-tion were closely coupled, an excellent demonstration of conservation of mass inEast Creek.5.4.4 Peak Discharge, Erosion, and DepositionThere were very low positive correlations between annual peak discharge and mag-nitude of bed material erosion/deposition in riffles and pools. This implies thatwhen using annual peak discharge as a predictor of erosion and deposition pat-terns, one must also consider additional factors governing transport processes, suchas sediment supply and bed conditioning. Additionally, the correlations betweenpeak discharge and mean bed deposition were consistently higher than those be-tween peak discharge and mean bed erosion (Tables 4.11 and 4.12). This differ-ence is difficult to explain given that erosion and deposition are tightly coupled inEast Creek (Subsection 5.4.3). However, it is distinctly possible that the correla-tions were so small that this difference had inconsequential physical significance,further emphasizing the importance of integrating multiple sediment transport gov-erning factors in analyzing erosion and deposition trends.75Chapter 6ConclusionsThe sediment transport and storage trends observed in East Creek at the channel,reach, and unit scales for WY04-11 offer useful insights into the state of the studyreach, factors governing sediment transport, and coupling of erosion and depositionprocesses.At the channel scale, relative stability in channel grade and elevation was ob-served over the study period, with localized areas of large bed elevation fluctuationsoccurring primarily around large wood and other channel obstructions.At the reach scale, there was considerable fluctuation in both bed load flux andstorage as obtained from the pit trap method and in bed material storage and fluxestimates as obtained from the morphological method. The pit trap method consis-tently underestimated storage compared to the morphological method as a result ofoverpassing of fine material and trap inefficiency. Both measurement methods esti-mated the same direction of storage (+ or -) in all cases. The highest magnitude bedload and bed material flux occurred during the year of the second highest annualpeak flow; however, other trends in bed load and bed material flux could not beexplained by flow regime. In-stream sediment supply conditions of the given andsurrounding reaches in the given and previous years helped to explain annual reachscale fluctuations in erosion and deposition. Large inputs of sediment in down-stream reaches (RP2, RP3) were repeatedly attributed to preceding evacuations ofsediment in upstream reaches (RAP, RP1). Large evacuations of sediment in down-stream reaches (RP2, RP3) repeatedly followed large inputs into these reaches in76previous years. Large increases in storage in upstream reaches (RAP, RP1) typi-cally followed large evacuations of sediment from these reaches in previous years.The somewhat balanced fluctuations in sediment storage in East Creek suggest astate of relative equilibrium within the study reach at the ten year scale of study.At the unit spatial scale, the detailed net bed elevation change, erosion, anddeposition histograms provided a unique opportunity to observe conservation ofmass, and coupling of erosion and deposition in East Creek. Fine scale bed eleva-tion changes varied across riffle and pools and could not be explained using flowregime and sediment supply alone. It is hypothesized that grain interactions mighthelp explain differences in spatial distributions of erosion and deposition. Therewas typically higher variance in elevation change, higher magnitudes of erosion,and higher magnitudes of deposition in pools compared to riffles. Taking graininteractions into account, this spatial heterogeneity in sediment storage at the unitscale can easily be explained. Pools contain finer more loosely interacting particlesthat are, consequently, more easily mobilized compared to those in riffles.Amassing the above conclusions, there are three salient points that can bedrawn from this thesis:1. In addition to flow and sediment supply regime, in-stream sediment supplyand bed conditioning also govern sediment transport processes.2. A detailed sediment budget using multiple measurement methods can havean amazing capacity to demonstrate conservation of mass and coupling oferosion and deposition.3. Explaining the physical significance of observed erosion and deposition trendsin a channel is tremendously aided by integrating spatial and temporal influ-ences at multiple scales.Additional research into sediment supply and storage that is spatially and tem-porally contextualized at multiple scales is recommended to continue unearthingthe enigma of sediment transport.77BibliographyJ. M. Buffington and D. R. Montgomery. A systematic analysis of eight decadesof incipient motion studies, with special reference to gravel-bedded rivers.Water Resources Research, 33(8):1993–2029, 1997. → pages 3, 4J. M. Buffington, R. D. Woodsmith, D. B. Booth, D. R. Montgomery, et al.Fluvial processes in Puget Sound rivers and the Pacific Northwest. Restorationof Puget Sound Rivers. University of Washington Press, Seattle, WA, pages46–78, 2003. → pages 2J. Caulkins. Sediment transport and channel morphology within a small forestedstream: East creek, British Columbia. Unpublished dissertation, University ofBritish Columbia, Vancouver. → pages 9, 12, 14, 16, 17M. Church. Bed material transport and the morphology of alluvial river channels.Annu. Rev. Earth Planet. Sci., 34:325–354, 2006. → pages 2, 8N. Clifford. Differential bed sedimentology and the maintenance of riffle-poolsequences. Catena, 20(5):447–468, 1993. → pages 3, 4G. A. M. de Almeida and J. F. Rodrı´guez. Understanding pool-riffle dynamicsthrough continuous morphological simulations. Water Resources Research, 47(1), 2011. → pages 2W. Dietrich, P. Nelson, E. Yager, J. Venditti, M. Lamb, and L. Collins. River,Coastal, and Estuarine Morphodynamics, Volume 1, Sediment patches,sediment supply, and channel morphology, pages 79–90. Taylor and Francis,2006. → pages 6B. Eaton, M. Hassan, and S. Davidson. Modeling wood dynamics, jam formation,and sediment storage in a gravel-bed stream. Journal of Geophysical Research:Earth Surface (2003–2012), 117(F4), 2012. → pages 5778P. Frey and M. Church. Bedload: a granular phenomenon. Earth SurfaceProcesses and Landforms, 36(1):58–69, 2011. → pages 7K. M. Frothingham, B. L. Rhoads, and E. E. Herricks. A multiscale conceptualframework for integrated ecogeomorphological research to support streamnaturalization in the agricultural midwest. Environmental Management, 29(1):16–33, 2002. → pages 2C. Garcia, J. B. Laronne, and M. Sala. Continuous monitoring of bedload flux in amountain gravel-bed river. Geomorphology, 34(1):23–31, 2000. → pages 6B. Gomez and M. Church. An assessment of bed load sediment transportformulae for gravel bed rivers. Water Resources Research, 25(6):1161–1186,1989. → pages 5D. G. Ham and M. Church. Bed-material transport estimated from channelmorphodynamics: Chilliwack river, British Columbia. Earth Surface Processesand Landforms, 25(10):1123–1142, 2000. → pages 8M. A. Hassan and I. Reid. The influence of microform bed roughness elements onflow and sediment transport in gravel bed rivers. Earth Surface Processes andLandforms, 15(8):739–750, 1990. → pages 5M. A. Hassan, M. Church, T. E. Lisle, F. Brardinoni, L. Benda, and G. E. Grant.Sediment transport and channel morphology of small, forested streams.Journal of the American Water Resources Association, 41(4):853–876, 2005.→ pages 1, 2, 3, 4M. A. Hassan, B. J. Smith, D. L. Hogan, D. S. Luzi, A. E. Zimmermann, andB. C. Eaton. Sediment storage and transport in coarse bed streams: scaleconsiderations. Developments in Earth Surface Processes, 11:473–496, 2007.→ pages 2, 8F. Iseya and H. Ikeda. Pulsations in bedload transport rates induced by alongitudinal sediment sorting: A flume study using sand and gravel mixtures.Geografiska Annaler. Series A. Physical Geography, 69:15–27, 1987. → pages7J. R. Jang and N. Gulley. Fuzzy logic toolbox users guide. The Mathworks Inc, 1:995, 1995. → pages 19D. Knighton. Fluvial forms and processes: a new perspective. Routledge, 2014.→ pages 279G. M. Kondolf and H. Pie´gay. Tools in fluvial geomorphology. Wiley OnlineLibrary, 2003. → pages 5H. Lamarre and A. G. Roy. A field experiment on the development of sedimentarystructures in a gravel-bed river. Earth Surface Processes and Landforms, 33(7):1064–1081, 2008. → pages 7M. A. Lenzi, V. D’Agostino, and P. Billi. Bedload transport in the instrumentedcatchment of the Rio Cordon: Part I: Analysis of bedload records, conditionsand threshold of bedload entrainment. Catena, 36(3):171–190, 1999. → pages5L. B. Leopold. A View of the River. Harvard University Press, 1994. → pages 5, 6T. E. Lisle, M. Madej, et al. Spatial variation in armouring in a channel with highsediment supply. Dynamics of Gravel-Bed Rivers, pages 277–293, 1992. →pages 2T. E. Lisle, J. M. Nelson, J. Pitlick, M. A. Madej, and B. L. Barkett. Variability ofbed mobility in natural, gravel-bed channels and adjustments to sediment loadat local and reach scales. Water Resources Research, 36(12):3743–3755, 2000.→ pages 6, 7R. J. McCleary and M. A. Hassan. Predictive modeling and spatial mapping offish distributions in small streams of the Canadian Rocky Mountain foothills.Canadian Journal of Fisheries and Aquatic Sciences, 65(2):319–333, 2008. →pages 1D. R. Montgomery and J. M. Buffington. Channel-reach morphology in mountaindrainage basins. Geological Society of America Bulletin, 109(5):596–611,1997. → pages 2, 3, 4P. A. Nelson, J. G. Venditti, W. E. Dietrich, J. W. Kirchner, H. Ikeda, F. Iseya, andL. S. Sklar. Response of bed surface patchiness to reductions in sedimentsupply. Journal of Geophysical Research: Earth Surface (2003–2012), 114(F2), 2009. → pages 6D. F. Oldmeadow and M. Church. A field experiment on streambed stabilizationby gravel structures. Geomorphology, 78(3):335–350, 2006. → pages 7C. Paola and R. Seal. Grain size patchiness as a cause of selective deposition anddownstream fining. Water Resources Research, 31(5):1395–1407, 1995. →pages 780G. Parker. Sedimentation Engineering, Volume 110, Transport of Gravel andSediment Mixtures, pages 165–251. American Society of Civil Engineers,2008. doi:10.1061/9780784408148.ch03. URL http://ascelibrary.org/doi/abs/10.1061/9780784408148.ch03. →pages 6R. L. Shreve. Stream lengths and basin areas in topologically random channelnetworks. The Journal of Geology, pages 397–414, 1969. → pages 1A. N. Strahler. Quantitative analysis of watershed geomorphology. Eos,Transactions American Geophysical Union, 38(6):913–920, 1957. → pages 1D. M. Thompson. The velocity-reversal hypothesis revisited. Progress inPhysical Geography, 35(1):123–132, 2011. → pages 4J. Venditti, W. Dietrich, P. Nelson, M. Wydzga, J. Fadde, and L. Sklar. Effect ofsediment pulse grain size on sediment transport rates and bed mobility in gravelbed rivers. Journal of Geophysical Research: Earth Surface (2003–2012), 115(F3), 2010. → pages 6J. M. Wheaton. Uncertainty in morphological sediment budgeting of rivers. PhDthesis, Centre for Ecology & Hydrology, 2008. → pages 19J. M. Wheaton, J. Brasington, S. E. Darby, and D. A. Sear. Accounting foruncertainty in DEMs from repeat topographic surveys: improved sedimentbudgets. Earth Surface Processes and Landforms, 35(2):136–156, 2010. →pages 19P. R. Wilcock. Toward a practical method for estimating sediment-transport ratesin gravel-bed rivers. Earth Surface Processes and Landforms, 26(13):1395–1408, 2001. → pages 5L. Wittenberg and M. D. Newson. Particle clusters in gravel-bed rivers: anexperimental morphological approach to bed material transport and stabilityconcepts. Earth Surface Processes and Landforms, 30(11):1351–1368, 2005.→ pages 7L. Wittenberg, J. Laronne, and M. Newson. Bed clusters in humid perennial andmediterranean ephemeral gravel-bed streams: The effect of clast size and bedmaterial sorting. Journal of Hydrology, 334(3):312–318, 2007. → pages 6B. Yuill, M. Nichols, and E. Yager. Coarse bed material patch evolution inlow-order, ephemeral channels. Catena, 81(2):126–136, 2010. → pages 6, 781A. Zimmermann and M. Church. Channel morphology, gradient profiles and bedstresses during flood in a step–pool channel. Geomorphology, 40(3):311–327,2001. → pages 382Appendix ASupporting Results: BedElevation ChangeAppendix A contains supporting figures and tables detailing bed elevation changepatterns in riffles and pools in East Creek from WY04-11. Appendix A includeshistograms (Figures A.1 - A.4), summary statistics (Tables A.1 - 4.6), and com-parisons of skew between riffles and pools (Figure A.5).83A.1 HistogramsYear Riffles Pools Year Riffles PoolsWY04 WY08WY05 WY09WY06 WY10WY07 WY11Figure A.1: Distribution of annual bed elevation changes in the riffles and pools within the rapids reach of East Creekfrom WY04 to WY1184Year Riffles Pools Year Riffles PoolsWY04 WY08WY05 WY09WY06 WY10WY07 WY11Figure A.2: Distribution of annual bed elevation changes in the riffles and pools within the riffle pool 1 sub-reach ofEast Creek from WY04 to WY1185Year Riffles Pools Year Riffles PoolsWY04 WY08WY05 WY09WY06 WY10WY07 WY11Figure A.3: Distribution of annual bed elevation changes in the riffles and pools within the riffle pool 2 sub-reach ofEast Creek from WY04 to WY1186Year Riffles Pools Year Riffles PoolsWY04 WY08WY05 WY09WY06 WY10WY07 WY11Figure A.4: Distribution of annual bed elevation changes in the riffles and pools within the riffle pool 3 sub-reach ofEast Creek from WY04 toWY1187A.2 Summary StatisticsTable A.1: Summary statistics for bed elevation change in riffles in rapidsreachYear Mean Median Variance Skew Kurtosis(m) (m) (m2)WY04 0.02 0.00 0.01 1.93 8.92WY05 -0.04 0.00 0.02 -2.96 10.96WY06 0.00 -0.01 0.01 3.66 24.04WY07 -0.01 0.00 0.01 -1.20 8.11WY08 -0.01 0.00 0.01 -2.26 17.72WY09 0.00 -0.01 0.01 0.74 4.91WY10 0.01 0.02 0.01 -0.27 4.00WY11 0.00 0.00 0.01 -0.46 9.51Average 0.00 0.00 0.01 -0.10 11.02Table A.2: Summary statistics for bed elevation change in riffles in riffle pool1 sub-reachYear Mean Median Variance Skew Kurtosis(m) (m) (m2)WY04 0.01 0.00 0.01 2.12 11.26WY05 0.00 0.01 0.01 -1.63 6.22WY06 -0.03 -0.02 0.00 -0.73 10.94WY07 0.00 -0.01 0.01 2.64 19.63WY08 -0.01 -0.01 0.00 0.02 12.21WY09 -0.01 0.00 0.01 -0.89 9.96WY10 0.01 0.01 0.00 1.21 20.61WY11 0.01 0.00 0.00 0.19 9.53Average 0.00 0.00 0.01 0.37 12.5588Table A.3: Summary statistics for bed elevation change in riffles in riffle pool2 sub-reachYear Mean Median Variance Skew Kurtosis(m) (m) (m2)WY04 -0.01 -0.01 0.00 1.04 4.43WY05 -0.02 -0.02 0.00 -0.15 1.77WY06 0.00 0.01 0.00 -1.28 13.58WY07 0.03 0.03 0.01 0.27 0.37WY08 -0.01 0.00 0.00 -4.76 40.25WY09 0.00 0.00 0.00 0.49 7.18WY10 -0.01 0.00 0.00 -1.68 7.38WY11 0.00 0.00 0.00 0.90 7.15Average 0.00 0.00 0.00 -0.64 10.26Table A.4: Summary statistics for bed elevation change in pools in rapidsreachYear Mean Median Variance Skew Kurtosis(m) (m) (m2)WY04 0.03 0.02 0.00 0.49 1.66WY05 -0.03 -0.03 0.00 0.79 5.27WY06 0.01 0.01 0.00 -0.83 5.10WY07 -0.01 -0.02 0.00 1.08 2.34WY08 0.01 0.01 0.00 0.31 6.26WY09 0.00 -0.01 0.00 0.75 5.39WY10 0.03 0.03 0.00 -1.47 7.43WY11 -0.01 -0.01 0.00 0.68 5.74Average 0.00 0.00 0.00 0.23 4.9089Table A.5: Summary statistics for bed elevation change in pools in riffle pool1 sub-reachYear Mean Median Variance Skew Kurtosis(m) (m) (m2)WY04 0.01 0.00 0.01 -0.04 7.45WY05 -0.05 -0.03 0.02 -1.15 5.14WY06 0.00 0.00 0.01 -0.09 5.98WY07 0.00 -0.01 0.01 0.49 3.12WY08 0.00 0.01 0.01 -0.77 5.94WY09 0.00 0.01 0.01 -0.97 11.88WY10 0.01 0.01 0.01 -0.31 10.41WY11 0.02 0.00 0.01 2.25 10.30Average 0.00 0.00 0.01 -0.07 7.53Table A.6: Summary statistics for bed elevation change in pools in riffle pool2 sub-reachYear Mean Median Variance Skew Kurtosis(m) (m) (m2)WY04 0.00 0.00 0.01 -0.52 6.64WY05 -0.06 -0.05 0.02 -0.41 1.12WY06 0.01 0.01 0.01 -0.28 7.08WY07 0.06 0.05 0.02 0.12 0.23WY08 0.00 0.00 0.01 -0.44 5.03WY09 0.02 0.01 0.01 1.27 7.68WY10 0.00 0.01 0.01 -1.46 9.10WY11 0.01 0.01 0.01 2.06 14.14Average 0.00 0.00 0.01 0.04 6.3890A.3 Skew-4-3-2-10123404 05 06 07 08 09 10 11Skewness Water Year (a) Rapids-2-1.5-1-0.500.511.522.5304 05 06 07 08 09 10 11Skewness Water Year (b) Riffle Pool 1-6-5-4-3-2-1012304 05 06 07 08 09 10 11Skewness Water Year (c) Riffle Pool 2-2-1.5-1-0.500.511.504 05 06 07 08 09 10 11Skewness Water Year (d) Riffle Pool 3Figure A.5: Skew in bed elevation over time in the riffle units (purple line) and the pool units (blue line) in the rapids,riffle pool 1, riffle pool 2, and riffle pool 3 sub-reaches of East Creek91Appendix BSupporting Results: Erosion andDepositionAppendix B contains supporting figures and tables detailing erosion and depositionpatterns in riffles and pools in East Creek from WY04-11. Appendix B includeshistograms (Figures B.1 - B.4) and summary statistics (Tables B.1 - B.12).92B.1 HistogramsYear Riffles Pools Year Riffles PoolsWY04 WY08WY05 WY09WY06 WY10WY07 WY11Figure B.1: Distribution of annual bed erosion and deposition in the riffles and pools in the rapids reach of East Creekfrom WY04 to WY1193Year Riffles Pools Year Riffles PoolsWY04 WY08WY05 WY09WY06 WY10WY07 WY11Figure B.2: Distribution of annual bed erosion and deposition in the riffles and pools within the riffle pool 1 sub-reachof East Creek from WY04 to WY1194Year Riffles Pools Year Riffles PoolsWY04 WY08WY05 WY09WY06 WY10WY07 WY11Figure B.3: Distribution of annual bed erosion and deposition in the riffles and pools within the riffle pool 2 sub-reachof East Creek from WY04 to WY1195Year Riffles Pools Year Riffles PoolsWY04 WY08WY05 WY09WY06 WY10WY07 WY11Figure B.4: Distribution of annual bed erosion and deposition in the riffles and pools within the riffle pool 3 sub-reachof East Creek from WY04 to WY1196B.2 Summary StatisticsTable B.1: Summary statistics for erosion in riffles in the rapids reachYear Mean Median Variance Skew Kurtosis(m) (m) (m2)WY04 -0.04 -0.03 0.00 -2.70 9.97WY05 -0.12 -0.06 0.03 -2.64 6.84WY06 -0.04 -0.03 0.00 -2.63 8.20WY07 -0.08 -0.05 0.01 -3.71 20.98WY08 -0.05 -0.03 0.01 -4.14 23.32WY09 -0.04 -0.03 0.00 -2.51 9.39WY10 -0.07 -0.05 0.00 -1.46 2.10WY11 -0.04 -0.03 0.00 -4.12 24.93Average -0.06 -0.04 0.01 -2.99 13.21Table B.2: Summary statistics for erosion in riffles in riffle pool 1 sub-reachYear Mean Median Variance Skew Kurtosis(m) (m) (m2)WY04 -0.04 -0.03 0.00 -2.81 10.25WY05 -0.06 -0.04 0.01 -2.57 7.15WY06 -0.05 -0.04 0.00 -3.15 25.18WY07 -0.06 -0.05 0.00 -1.07 2.15WY08 -0.04 -0.03 0.00 -3.48 18.87WY09 -0.05 -0.03 0.00 -3.07 13.16WY10 -0.04 -0.02 0.00 -2.88 10.83WY11 -0.03 -0.02 0.00 -3.99 22.35Average -0.05 -0.03 0.00 -2.88 13.7597Table B.3: Summary statistics for erosion in riffles in riffle pool 3 sub-reachYear Mean Median Variance Skew Kurtosis(m) (m) (m2)WY04 -0.03 -0.02 0.00 -3.24 17.78WY05 -0.04 -0.03 0.00 -2.31 6.93WY06 -0.02 -0.02 0.00 -2.59 10.98WY07 -0.04 -0.03 0.00 -1.66 5.25WY08 -0.03 -0.02 0.00 -4.14 26.87WY09 -0.02 -0.02 0.00 -3.41 19.92WY10 -0.03 -0.02 0.00 -2.31 8.43WY11 -0.02 -0.02 0.00 -2.83 12.14Average -0.03 -0.02 0.00 -2.81 13.54Table B.4: Summary statistics for deposition in riffles in the rapids reachYear Mean Median Variance Skew Kurtosis(m) (m) (m2)WY04 0.08 0.04 0.01 2.68 9.74WY05 0.05 0.04 0.00 1.91 5.44WY06 0.05 0.02 0.01 4.06 17.86WY07 0.07 0.05 0.00 1.61 3.37WY08 0.04 0.03 0.00 3.46 18.49WY09 0.05 0.03 0.00 2.36 7.09WY10 0.05 0.04 0.00 2.98 13.60WY11 0.05 0.03 0.00 2.34 6.83Average 0.06 0.03 0.00 2.68 10.3098Table B.5: Summary statistics for deposition in riffles in riffle pool 1 sub-reachYear Mean Median Variance Skew Kurtosis(m) (m) (m2)WY04 0.06 0.03 0.01 3.17 12.68WY05 0.05 0.04 0.00 1.75 4.69WY06 0.03 0.02 0.00 3.36 14.83WY07 0.06 0.05 0.01 5.38 36.95WY08 0.04 0.02 0.00 3.88 20.12WY09 0.04 0.02 0.00 3.63 17.08WY10 0.04 0.02 0.00 5.78 46.25WY11 0.04 0.03 0.00 2.89 11.68Average 0.04 0.03 0.00 3.73 20.53Table B.6: Summary statistics for deposition in riffles in riffle pool 3 sub-reachYear Mean Median Variance Skew Kurtosis(m) (m) (m2)WY04 0.03 0.02 0.00 2.73 10.63WY05 0.03 0.02 0.00 2.69 11.59WY06 0.03 0.02 0.00 2.64 9.44WY07 0.06 0.06 0.00 1.67 4.15WY08 0.02 0.02 0.00 3.35 17.85WY09 0.03 0.02 0.00 2.38 9.11WY10 0.02 0.01 0.00 2.75 12.43WY11 0.03 0.02 0.00 3.40 16.46Average 0.03 0.02 0.00 2.70 11.4699Table B.7: Summary statistics for erosion in pools in rapids reachYear Mean Median Variance Skew Kurtosis(m) (m) (m2)WY04 -0.03 -0.02 0.00 -1.70 3.48WY05 -0.06 -0.04 0.00 -1.33 1.86WY06 -0.04 -0.02 0.00 -2.54 8.02WY07 -0.05 -0.05 0.00 -0.83 0.74WY08 -0.03 -0.02 0.00 -2.17 6.03WY09 -0.03 -0.02 0.00 -2.73 10.44WY10 -0.06 -0.04 0.00 -1.96 3.81WY11 -0.03 -0.02 0.00 -1.90 5.15Average -0.04 -0.03 0.00 -1.89 4.94Table B.8: Summary statistics for erosion in pools in riffle pool 1 sub-reachYear Mean Median Variance Skew Kurtosis(m) (m) (m2)WY04 -0.07 -0.04 0.01 -3.48 16.14WY05 -0.10 -0.07 0.01 -2.36 7.10WY06 -0.05 -0.03 0.00 -2.84 10.28WY07 -0.08 -0.05 0.01 -1.67 3.01WY08 -0.08 -0.03 0.01 -2.05 5.38WY09 -0.05 -0.03 0.01 -3.26 14.30WY10 -0.05 -0.02 0.00 -2.86 10.81WY11 -0.04 -0.02 0.00 -3.43 19.06Average -0.06 -0.04 0.01 -2.75 10.76100Table B.9: Summary statistics for erosion in pools in riffle pool 3 sub-reachYear Mean Median Variance Skew Kurtosis(m) (m) (m2)WY04 -0.08 -0.06 0.01 -2.15 6.72WY05 -0.09 -0.07 0.01 -1.69 3.85WY06 -0.05 -0.03 0.00 -2.84 9.87WY07 -0.05 -0.04 0.00 -2.79 14.75WY08 -0.06 -0.04 0.01 -2.84 11.94WY09 -0.05 -0.03 0.00 -2.89 11.25WY10 -0.06 -0.03 0.01 -2.23 5.59WY11 -0.05 -0.03 0.00 -2.39 9.16Average -0.06 -0.04 0.00 -2.48 9.14Table B.10: Summary statistics for deposition in pools in rapids reachYear Mean Median Variance Skew Kurtosis(m) (m) (m2)WY04 0.05 0.04 0.00 1.48 2.53WY05 0.04 0.03 0.00 3.24 15.44WY06 0.03 0.02 0.00 1.88 3.79WY07 0.06 0.04 0.00 1.78 2.98WY08 0.04 0.03 0.00 3.10 13.98WY09 0.04 0.03 0.00 2.45 6.88WY10 0.04 0.03 0.00 1.67 2.80WY11 0.03 0.02 0.00 2.82 13.38Average 0.04 0.03 0.00 2.30 7.72101Table B.11: Summary statistics for deposition in pools in riffle pool 1 sub-reachYear Mean Median Variance Skew Kurtosis(m) (m) (m2)WY04 0.08 0.05 0.01 2.44 7.66WY05 0.06 0.04 0.01 2.47 8.01WY06 0.05 0.03 0.00 2.44 8.80WY07 0.09 0.06 0.01 2.18 6.68WY08 0.06 0.04 0.00 2.95 12.14WY09 0.05 0.03 0.00 3.45 18.42WY10 0.04 0.03 0.00 3.59 18.25WY11 0.07 0.03 0.01 2.69 9.08Average 0.06 0.04 0.01 2.78 11.13Table B.12: Summary statistics for deposition in pools in riffle pool 3 sub-reachYear Mean Median Variance Skew Kurtosis(m) (m) (m2)WY04 0.06 0.05 0.00 1.94 5.62WY05 0.09 0.05 0.01 1.34 0.96WY06 0.04 0.02 0.00 2.96 11.73WY07 0.08 0.06 0.01 1.69 4.21WY08 0.05 0.03 0.00 3.01 11.86WY09 0.06 0.04 0.00 2.64 9.28WY10 0.03 0.03 0.00 2.66 13.12WY11 0.06 0.04 0.00 2.66 9.83Average 0.06 0.04 0.00 2.36 8.33102

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0167165/manifest

Comment

Related Items