UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Temporal patterns in the normal-regime fine-sediment cascade in Russell Creek Basin, Vancouver Island Nistor, Craig 1996

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

Item Metadata

Download

Media
831-ubc_1996-0096.pdf [ 17.15MB ]
Metadata
JSON: 831-1.0087011.json
JSON-LD: 831-1.0087011-ld.json
RDF/XML (Pretty): 831-1.0087011-rdf.xml
RDF/JSON: 831-1.0087011-rdf.json
Turtle: 831-1.0087011-turtle.txt
N-Triples: 831-1.0087011-rdf-ntriples.txt
Original Record: 831-1.0087011-source.json
Full Text
831-1.0087011-fulltext.txt
Citation
831-1.0087011.ris

Full Text

T E M P O R A L P A T T E R N S I N T H E N O R M A L - R E G I M E F I N E - S E D I M E N T C A S C A D E I N R U S S E L L C R E E K B A S I N , V A N C O U V E R I S L A N D by C R A I G N I S T O R B.Sc : ,The University of British Columbia, 1990 A T H E S I S S U B M I T T E D I N P A R T I A L F U L F I L L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F M A S T E R O F S C I E N C E in T H E F A C U L T Y OF G R A D U A T E S T U D I E S (Department of Geography) We accept this thesis as conforming to the required standard T H E U N I V E R S I T Y OF B R I T I S H C O L U M B I A January 1996 © Craig JohnNistor, 1996 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of GkoGK-fiPHy The University of British Columbia Vancouver, Canada Date °l /U DE-6 (2/88) ABSTRACT Large, infrequent "episodic" sediment transfers are commonly considered differently from "normal-regime" sediment-transfer activity. For example, in the important hillslope-gully-stream sediment cascade pathway in coastal British Columbia, debris slides and debris torrents are considered as "episodic events". On the other hand, lower-magnitude hillslope to gully-channel sediment transfers and fluvial sediment tranport within gully and stream channels are usually considered as "normal-regime" activity, represented by annual yields. However, the results of this study illustrate the highly episodic nature of normal-regime fine-sediment transfers, which are closely linked to hydrometeorological and sediment-supply conditions. The results indicate that qualitative modelling of fine-sediment transfer activity, at the synoptic or event scale, should be possible based upon hydrometeorological and sediment-supply information. From such a model ~ the elements of which are presented in the concluding chapter ~ fine-sediment transfer activity could be forecast based upon regional weather forecasts. The study was conducted in Russell Creek Basin, on northern Vancouver Island, British Columbia. Fine-sediment transfer activity was monitored at a nested hierarchy of sites representing fine-sediment transfers from unstable hillslopes to a gully channel, suspended sediment transport out of the unstable gully and a nearby stable gully, and suspended sediment transport in Russell Creek near the mouth. Russell Creek Basin is located within Tsitika Watershed, which is the site of a British Columbia Ministry of Forests study dedicated to determining relative fine-sediment contributions from natural and logging-related sediment sources. The results of the Russell Creek study indicate that an event-based sediment sampling program is desirable and that at least some automated sampling is required. Furthermore, development of a qualitative sediment-transfer ii activity forecast model would be useful in interpretation of sample data and would allow efforts to be concentrated during the periods of greatest sediment-transfer activity. ( TABLE OF CONTENTS Page Abstract Table of Contents List of Tables List of Figures Acknowledgements Chapter 1. Introduction and Overview 1 1.1 Introduction 1 1.2 Sediment cascade scales : the study in perspective 3 1.2.1 Sediment-cascade scale relations 4 1.2.2 Classification of scale relations 5 1.2.3 Sediment cascade scales considered in this study 6 1.3 Normal-regime fine-sediment cascade in coastal British Columbia 7 1.3.1 Landform units in coastal British Columbia 7 1.3.2 Hillslope to gully-channel sediment transfers 8 1.3.3 Fluvial sediment transport in gully channels 17 1.3.4 Fluvial sediment transport in streams 20 1.4 Short-term temporal patterns in fine-sediment transfers 22 1.4.1 Hydrometeorological conditions 22 1.4.2 Hillslopes to gully channels 24 1.4.3 Gullies and streams to basin outlet 25 iv ii iv x xii xvii 1.5 Land management and temporal patterns in the fine-sediment cascade 31 1.5.1 Forestry and fine sediment 32 1.5.2 Fine sediment and salmonid habitat 34 1.5.3 Significance of temporal patterns in fine-sediment transfers 35 1.6 Study hypotheses 36 1.6.1 General hypothesis 36 1:6.2 Specific hypotheses 37 Chapter 2. Study Setting and Design 41 2.1 Nested spatial sediment-monitoring scales 43 2.2 Hydrometeorological conditions 43 2.3 Physical characteristics of Russell Creek Basin 46 2.3.1 Physiography 47 2.3.2 Climate 49 2.3.3 Vegetation and land use 50 2.3.4 Russell Creek channel 50 2.4 Physical characteristics of the study gullies 50 2.4.1 Gully situation 53 2.4.2 Gully morphology 53 2.4.3 Sediment sources 54 2.4.4 Channel morphology 55 2.5 Study sediment sources 58 2.6 Summary 60 Chapter 3. Methods . 61 3.1 Hydrometeorological conditions 61 3.1.1 Weather 61 3.1.2 Snowpack depth 62 3.1.3 Synoptic-scale weather system classification 62 3.2 Hillslope to gully-channel sediment transfers 63 3.2.1 Erosion / deposition pins 63 3.2.2 Sediment trapping 68 3.3 Suspended sediment transport in gullies 68 3.3.1 Gully gauging stations 68 3.3.2 Stage and discharge 69 3.3.3 Turbidity and suspended sediment concentration 75 3.4 Suspended sediment transport in Russell Creek 80 Chapter 4. Hydrometeorological Conditions and Hillslope to Gully-Channel Sediment Transfers 83 4.1 Hydrometeorological conditions 83 4.1.1 Observations 83 4.1.2 Weather in Russell Creek Basin compared to Port Hardy Airport 87 4.1.3 Study-period weather in perspective 91 4.1.4 Runoff events in Russell Creek compared to Tsitika River 93 4.2 Hillslope to gully-channel sediment transfers 96 4.2.1 Sediment transfer activity at the monitored sites 96 4.2.2 Hillslope sediment source activity and hydrometeorological conditions 100 Chapter 5. Suspended Sediment Transport in Gullies 102 5.1 Hydrological comparison of Gullies B and C 104 vi 5.2 Importance of runoff events in transporting suspended sediment 108 5.3 Seasonal scale patterns . 110 5.4 Sub-seasonal scale patterns 114 5.4.1 Gauging Station B .114 5.4.2 Gauging Station C 119 5.4.3 Summary 121 5.5 Synoptic scale patterns 126 5.5.1 Runoff Event 1 at Gauging Station C 126 5.5.2 Runoff Event 2 at Gauging Station B 129 5.5.3 Runoff Event 6 at Gauging Stations B and C 132 5.5.4 RunoffEvent 7 at Gauging Station B 135 5.5.5 RunoffEvent 11 at Gauging Stations B and C 137 5.5.6 RunoffEvent 13 at Gauging Stations B and C 141 5.5.7 RunoffEvent 15 at Gauging Station B 144 5.5.8 Summary 146 5.6 Debris-torrent transport in Gully B 147 Chapter 6. Suspended Sediment Transport in Russell Creek 151 6.1 Hydrological comparison of Russell Creek Gauging Station to gully gauging stations 154 6.2 Importance of runoff events in transporting suspended sediment 154 6.3 Seasonal scale patterns 158 6.4 Synoptic scale patterns 162 6.4.1 Runoff Event 6 at Russell Creek Gauging Station 164 6.4.2 Runoff Event 8 at Russell Creek Gauging Station 164 6.4.3 Runoff Event 11 at Russell Creek Gauging Station 168 6.4.4 Summary 170 Chapter 7. Discussion and Conclusions 171 7.1 Discussion of hypotheses 171 7.1.1 Hillslope to gully-chatinel transfers 171 7.1.2 Suspended sediment transport in gullies: seasonal and sub-seasonal patterns 172 7.1.3 Suspended sediment transport in gullies: synoptic scale patterns 174 7.1.4 Suspended sediment transport in streams: seasonal and sub-seasonal patterns 176 7.1.5 Suspended sediment transport in streams: synoptic scale patterns 177 7.1.6 Debris-torrent versus fluvial fine-sediment transport 178 7.1.7 Summary: temporal characteristics of normal-regime fine-sediment transfers 178 7.2 Application to suspended sediment sampling programs 178 7.3 Elements of a normal-regime fine-sediment transfer activity model 180 7.4 Conclusions 183 References 185 Appendices Appendix A. Sediment transfer data A-1. Erosion / deposition pin data 192 A-2. Discharge and suspended sediment concentration at Gauging Station B 193 A-3. Discharge and suspended sediment concentration at Gauging Station C 206 viii A-4. Discharge and suspended sediment concentration at Russell Creek Gauging Station (WSC 08HF007) 217 Appendix B. Gully gauging station instrument calibrations -B- l . Stage to weir-notch head conversions 224 B-2. Stage-discharge rating formulae 226 B-3. Turbidity-SSC relations 229 Appendix C. Survey data C - l . Gully morphology surveys 231 C-2. Particle-size distributions of bulk-sampled sediment sources 233 C-3. Gully gauging station site sketches 235 IX LIST OF TABLES Table Page 1.1 Normal-regime fine-sediment transfer processes in the hillslope to gully-channel component of the sediment cascade 9 1.2 Literature summary: Seasonal and sub-seasonal patterns of fine-sediment transfers in small, normal-regime catchments 26 1.3 Literature summary: Synoptic scale (single-event) patterns of suspended sediment transport in small, normal-regime catchment channels 30 2.1 Physical characteristics of Gullies B and C 51 4.1 Summary of hydrometeorological conditions, Russell Creek Basin, September 1993 - June 1994 84 4.2 Hydrometeorological conditions summarised by seasons and sub-seasons 86 4.3 Monitored erosion / deposition at Gully B sediment source sites 97 5.1 Summary of data collection at Gauging Stations B and C 104 5.2 Summary of runoff events at Gauging Stations B and C 105 5.3 Importance of runoff events in transporting suspended sediment at Gauging Stations B and C 109 5.4 Synoptic scale characteristics of suspended sediment transport during runoff events at the gully gauging stations 127 6.1 Summary of data collection at Russell Creek Gauging Station 151 6.2 Summary of runoff events at Russell Creek Gauging Station 153 6.3 Importance of runoff events in transporting suspended sediment at Russell Creek Gauging Station 157 6.4 Synoptic-scale characteristics of suspended sediment transport during runoff events at Russell Creek Gauging Station LIST OF FIGURES Figure Page IT Idealised coastal British Columbia gully morphology 10 2.1 Location map, Russell Creek Basin 42 2.2 Suspended sediment gauging stations in Russell Creek Basin 44 2.3 Hydrometeorological stations and suspended sediment gauging stations in the vicinity of the study gullies 45 2.4 Russell Creek Basin, looking eastward from Tsitika Valley 47 2.5 View of study gullies 48 2.6 Gully B hillslope-to-channel sediment sources and gully long-profile 52 2.7 Gully C morphology 55 2.8 Gully B sidewall failures 56 2.9 Gully B active sidewall slope and road cutbank 57 2.10 The big L O D jam and associated sediment wedge in Gully B 59 3.1 Erosion / deposition pins 65 3.2 An example of needle ice growth in road cutbank sediment in Russell Creek Basin 67 3.3 Gauging Station B 70 3.4 Gauging Station C 71 3.5 Head-discharge rating curves for Gauging Stations B and C 76 3.6 Relations between turbidity and SSC at Gauging Stations B and C 79 3.7 Relation between turbidity and SSC at Russell Creek Gauging Station 82 4.1 Hydrometeorological conditions in Russell Creek Basin, September 1993 -June 1994 85 xii 4.2 Comparison of air temperatures at weather stations Russ2 and Port Hardy Airport 88 4.3 Comparison of daily rainfall at weather stations Russ2 and Port Hardy Airport 89 4.4 Mean monthly precipitation and temperature at Port Hardy Airport during the study period compared to long-term average 92 4.5 Comparison of peak runoff-event discharges in Russell Creek and Tsitika River 94 5.1 Discharge and suspended sediment concentration at Gauging Stations B and C, September 1993 -June 1994 103 5.2 Hydrological comparison of Gullies B and C 106 5.3 Areally-normalised hydrological comparison of Gullies B and C 107 5.4 Comparison of peak suspended sediment concentration to peak discharge during study-period runoff events at Gauging Stations B and C 111 5.5 Comparison of suspended sediment transport to runoff volume during study-period runoff events at Gauging Stations B and C 112 5.6 Event-by-event display of runoff volume and suspended sediment transport during study-period runoff events at Gauging Stations B and C 113 5.7 Comparison of suspended sediment transport to peak discharge during study-period runoff events at Gauging Stations B and C 115 5.8 Comparison of suspended sediment transport to runoff volume above threshold peak discharge (270 l/s) during large runoff events at Gauging Station B 116 5.9 Influence of peak discharge on suspended sediment transport during runoff events at Gauging Station B 117 5.10 Influence of peak discharge on suspended sediment transport during runoff events at Gauging Station C 120 5.11 Comparison of suspended sediment transport to runoff volume during runoff events categorised by peak discharge and sub-season 122 5.12 Comparison of suspended sediment transport to runoff volume during study-period runoff events at Gauging Stations B and C; sediment supply regimes identified 123 5.13 Comparison of specific suspended sediment transport to specific runoff volume during study-period runoff events at Gauging Stations B and C; sediment supply regimes identified 125 5.14 Discharge and suspended sediment concentration at Gauging Station C during RunoffEvent 1, 21-24 October, 1993 128 5.15 Discharge and suspended sediment concentration at Gauging Station B during Runoff Event 2, 2 -3 November, 1993 131 5.16 Discharge and suspended sediment concentration at Gauging Station B during RunoffEvent 6, 2 - 3 December, 1993 133 5.17 Discharge and suspended sediment concentration at Gauging Station C during Runoff Event 6, 2- 3 December, 1993 134 5.18 Discharge and suspended sediment concentration at Gauging Station B during RunoffEvent 7, 9 - 10 December, 1993 136 5.19 Discharge and suspended sediment concentration at Gauging Station B during RunoffEvent 11, 27 February-3 March, 1994 138 5.20 Discharge and suspended sediment concentration at Gauging Station C during RunoffEvent 11, 27 February-3 March, 1994 139 5.21 Discharge and suspended sediment concentration at Gauging Station B during RunoffEvent 13, 27 March - 1 April, 1994 142 xiv 5.22 Discharge and suspended sediment concentration at Gauging Station C during RunoffEvent 13, 27 March - 1 April, 1994 143 5.23 Discharge and suspended sediment concentration at Gauging Station B during RunoffEvent 15, 10 - 12 April, 1994 145 5.24 Gully B debris-torrent deposit, November 1994 149 6.1 Discharge and suspended sediment concentration at Russell Creek Gauging Station, September 1993 - June 1994 152 6.2 Hydrological comparison of Russell Creek Gauging Station to Gauging Stations B a n d C 155 6.3 Areally-normalised hydrological comparison of Russell Creek Gauging Station to Gauging Stations B and C 156 6.4 Event-by-event display of flow and suspended sediment transport characteristics during study-period runoff events at Russell Creek Gauging Station 159 6.5 Comparison of suspended sediment transport to peak discharge during study-period runoff events at Russell Creek Gauging Station 160 6.6 Comparison of runoff volume and suspended sediment transport during study-period runoff events at Russell Creek Gauging Station and Gauging Stations B and C 161 6.7 Discharge and suspended sediment concentration at Russell Creek Gauging Station during RunoffEvent 6, 2 - 3 December, 1993 165 6.8 Comparison of lab-calibrated SSC to sampled SSC at Russell Creek Gauging Station during RunoffEvent 6, 2 - 3 December, 1993 166 6.9 Discharge and suspended sediment concentration at Russell Creek Gauging Station during RunoffEvent 8, 29 December, 1993 - 5 January, 1994 167 6.10 Discharge and suspended sediment concentration at Russell Creek Gauging Station during RunoffEvent 11, 27 February - 3 March, 1994 ACKNOWLEDGEMENTS I owe thanks to a very large number of people. My supervisor, Dr. Michael Church, and my second reader, Dr. Michael Bovis, have both provided me not only with assistance in writing this thesis, but also with broader academic mentorship throughout my many years at UBC. British Columbia Ministry of Forests provided funding for this study as part of the Tsitika Watershed Sedimentation Study. Special thanks go to Mr. Dan Hogan and Dr. Rob Hudson of Ministry of Forests for support during this project. Thanks also to Mr. Ken Rood of Northwest Hydraulic Consultants Ltd. for support and patience during the final months of thesis writing. Field assistance was provided by numerous people, but greatest thanks must go to Darren Ham, who made several unpaid trips to Russell Creek, despite being fed raw hot-dogs in the pouring rain. Darren also has the distinction of having been with me on the last field trip ever made by UBC Geography's Dodge "Pig". Scott Davidson deserves special thanks for his assistance and readily proffered advice on our many field trips together. Thanks also go to Bobby Downs, Todd Williston, Ginnie Cosgrove, Steve Bird, Hjalmar Laudon and Asa, Magdalena Rucker, J.F. Proulx, Scott Weston, Brett Eton, and Laris Grikis. Thanks to MacMillan-Bloedel personnel at the Eve River Division for providing much useful information and a place to park our trailer and have a shower — a valuable gift, indeed. Finally, thanks to the management and patrons of the Smiling Salmon Pub in Sayward for providing many interesting experiences. I could not have made it to this point without the moral support provided by family and friends. Thank you, Mom and Dad and Jen. Thank you, David Jan and Gord Bradley, for your ongoing support, even if my consumption of your tax dollars was your main motivation. Thank you, Dr. Michael Pugsley, for providing an invaluable role model. Most importantly, thank you, Ronda, for your continual patience and inspiration and your seemingly endless faith in me. xvii Chapter 1 Introduction and Overview 1.1 Introduction Geomorphic activity occurs over a wide spectrum of magnitudes and frequencies. In coastal British Columbia, "normal-regime" activity, consisting of relatively low-magnitude high-frequency events, is considered to transfer minor, "background" quantities of sediment through the sediment cascade. In contrast, "episodic" events of relatively high magnitude and low frequency ~ such as landslides and debris torrents — occasionally punctuate the normal regime, transporting large volumes of sediment and altering the landscape during the course of single events. As a result, studies of episodic geomorphic activity usually consider individual events. Studies of normal-regime activity, on the other hand, usually consider generalised sediment transfer mechanisms and integrated sediment transfer volumes, such as annual surface erosion rates and annual basin sediment yields. Despite these methodological differences, normal-regime geomorphic activity also consists of discrete sediment-transfer events, occurring in smaller magnitudes and at higher spatial and/or temporal frequencies than episodic events. Occurrence of normal-regime events varies at temporal scales which reflect the dominant short-term frequencies in hydrometeorological conditions: synoptic, sub-seasonal and seasonal. Better understanding of sediment transfers at these temporal scales may improve the estimation of longer-term yields by uncovering flaws in 1 assumptions underlying average yield estimates and suggesting better assumptions to replace them. Understanding the normal-regime range of geomorphic activity is important because of its predominance in space and time, and its susceptibility to alteration by land management practices. Alteration of the sediment cascade in coastal British Columbia by land management activities, particularly forestry, has received considerable attention in recent decades. Alteration of the fine-fraction sediment cascade has been of particular interest where stream habitat and domestic water supplies are affected. Normal-regime sediment transfers preferentially involve fine-textured sediments because of the relatively low erosive energy involved. Knowledge of the "natural-state" of normal-regime geomorphic activity is needed in order to appraise the effects of land management on the fine-sediment cascade. General objectives This study was conducted to provide more information on the normal-regime fine-sediment cascade in coastal British Columbia. Particular attention was paid to temporal patterns of fine sediment transfers at synoptic, sub-seasonal and seasonal temporal scales. The study was conducted in Russell Creek Basin, tributary to Tsitika River on northern Vancouver Island. This study is part of the larger Tsitika Watershed Sedimentation Study being conducted by the British Columbia Ministry of Forests (Hogan and Chatwin 1990). The general objectives of the Russell Creek fine-sediment cascade study are as follows: 1) To monitor fine-sediment transfers at three nested spatial scales, representing a typical pathway through the Russell Creek Basin sediment cascade: a) from gully walls into gully channels by a variety of transfer mechanisms, b) out of gullies into Russell Creek by fluvial transport, c) out of Russell Creek basin by fluvial transport. 2) To characterise the temporal patterns of the fine-sediment transfers at hydrometeor-ologically controlled temporal scales ~ synoptic, sub-seasonal and seasonal — for one year. 3) To compare the temporal patterns between two gully types ~ gullies with unstable versus stable sidewalls ~ to gain some understanding of spatial variability in the landscape. 4) To characterise the study-period hydrometeorological conditions relative to long-term average conditions in order to assess the representativeness of the study. 5) To provide the elements of a qualitative model capable of forecasting relative activity of normal-regime fine-sediment transfers at the synoptic, sub-seasonal and seasonal temporal scales based upon regional hydrometeorological forecasts. This chapter provides an overview of temporal patterns of normal-regime fine-sediment cascades in coastal British Columbia and similar environments. A more specific set of study objectives and hypotheses follows at the end of the chapter. 1.2 Sediment cascade scales : the study in perspective A sediment cascade is the series of sediment transfers by which sediment moves from the land surface of a drainage basin into the basin's stream channel network, and thence to the basin's outlet. The sediment transfers involved in sediment cascades cover very great ranges of scale in four dimensions: time, space, magnitude, and sediment texture. This study aims to characterise a specific set of these scales; namely, the short-term temporal patterns of normal-regime, fine-sediment transfers through a small, coastal British Columbia drainage basin. 3 1.2.1 Sediment-cascade scale relations The dimensions of space, time, and magnitude in sediment transfers are closely related. The temporal and spatial frequencies of sediment transfers tend to be inversely related to transfer magnitudes (Wolman and Miller 1960). In other words, processes which transfer large amounts of sediment in single events — such as landsliding — tend to occur relatively infrequently in time and space, as compared to smaller magnitude transfer processes ~ such as dry ravel of streambanks ~ which occur relatively more frequently. Similarly, large events generated by a given process tend to occur less frequently in time and space than do small events generated by the same process (e.g., fewer large landslides than small landslides). Large sediment transfer events occur less frequently than do small events because successively larger events require successively greater mass and energy inputs delivered in successively more specific combinations of antecedent events, all of which are increasingly rare. The fourth dimension, sediment texture, is positively related to event magnitude, and inversely related to temporal and spatial frequency, in general. In other words, the average textural composition of sediment transferred coarsens, and the range of particle sizes involved increases, as transfer event magnitude increases. Since coarse sediment is made up of individually more massive particles than is fine sediment, with correspondingly higher thresholds of energy required to initiate motion, coarse sediment tends to be transferred more episodically than finer material. 1.2.2 Classification of scale ranges Because of the great range of scales of the four dimensions discussed above, and because of the general relations between the scales, it is useful to categorise sediment transfers into more manageable ranges of scales. Sediment transfer processes can be categorised as occurring "episodically" or in "normal regime", based on temporal and spatial frequency and event magnitude. Normal regime describes the state in which a drainage basin exists most of the time. In normal regime, low-magnitude high-frequency sediment transfers operate in a landscape which changes very slowly and steadily; indeed, the landscape may be considered to be in a steady-state in the short term (Schumm and Lichty 1965). When the normal regime is interrupted by episodic sediment transfer events, large volumes of sediment are transferred and the landscape is notably altered. The separation of episodic events from the normal regime is arbitrary. The main factors controlling the categorisation are length of time and spatial extent considered. Over increasingly long durations and increasingly broad areas, increasingly large and rare sediment transfer events must be considered a part of the normal regime. Conversely, over decreasingly short durations and decreasingly small areas, geomorphic activity previously considered as continuous is seen to be composed of discrete, episodic events. Sediment texture scales can be conveniently categorised based on dominant modes of transfer. Many processes transfer wide ranges of sediment particle sizes, with increasingly finer particles dominating the mixture in increasingly small, frequent events. However, sediment transport by flowing water is both a very common and a very texture-selective transfer process. Sediment particles less than about 0.18 mm in diameter are almost always transported in suspension in flowing water, while particles coarser than about 1 mm generally travel in traction along the streambed. Particles with diameters between 0.18 and 1 mm can be transported by 5 either process, but they are usually included in the suspended load during high flows. Thus, particle diameter of 1 mm is a useful division between "fine" and "coarse" sediment (Church, in press). 1.2.3 Sediment cascade scales considered in this study This study is concerned with the short-term patterns of normal-regime, fine-sediment transfers through small, coastal British Columbia drainage basins. To quantify, the maximum temporal duration and spatial extent of interest are on the order of one year and tens of square kilometres, respectively. One year constitutes a single cycle of seasonally varying hydrometeorological conditions which provide the energy for normal-regime sediment transfers. The spatial limit is based on the area of a typical small, coastal British Columbia drainage basin. The minimum time and space scales considered are on the order of one hour and one square metre, respectively. These minimum scales represent the temporal resolution of sediment transfer monitoring performed in this study and the area of individual sediment transfer sites. The temporal and spatial scale ranges involved are on the order of 10 4 and 108, respectively. The range of sediment transfer magnitudes considered is on the order of IO9 Minimum sediment transfer magnitudes are defined by measurement detectability and are probably on the order of cubic millimetres. At the other end of the scale, maximum sediment transfer magnitudes in normal regime are on the order of cubic metres; larger transfers constitute episodic events. In this study only transfers of fine sediment have been considered; namely, sediment particles less than 1 mm in diameter, ranging down to the minimum colloidal particle size of about 0.1 microns. This represents a particle diameter range on the order of 10 4 Thus, normal-regime, fine-sediment transfer processes in the context of this study are those that operate frequently throughout a small drainage basin during the course of a typical year. Basin characteristics such as morphometry and vegetation cover will be assumed to be constant in the short-term normal-regime study. Even within these seemingly restrictive study limits, however, the range of scales considered is large. 1.3 Normal-regime fine-sediment cascade in coastal British Columbia Sediment cascade pathways can be subdivided into components, each component consisting of the set of processes operating to transfer sediment from one landform unit to another. Numerous sediment cascade pathways exist, composed of various sets of sediment-transfer processes operating between various landform units (Church 1983). The dominant fine-sediment cascade pathway during normal regime in coastal British Columbia consists of three components: sediment transfers from hillslopes to gully channels, from gully channels to trunk stream channels, and along trunk stream channels to basin outlets. 1.3.1 Landform units in coastal British Columbia The landscape of coastal British Columbia is mountainous, with deep, smooth-sided, U-shaped, glacially-scoured valleys. This landscape can be classified into four characteristic landform units: hillslopes, gullies, toe-slopes and valley-flats, and stream channels. Hillslopes in coastal British Columbia are typically covered with veneers of glacial till and colluvium, and dissected by roughly parallel gullies. Gullies are narrow, steep, linear depressions, incised into the hillslopes, which contain first- and second-order tributary channels. Gullies drain into third- and higher-order channels, hereafter referred to simply as stream channels. Stream channels are typically gravel- or boulder-bedded, exhibit riffle-pool or step-pool morphology, and are at least somewhat buffered from direct hillslope influence by toe-slopes and valley-flats (Church 1992). Toe-slopes and valley-flats consist of colluvial and alluvial fans, and minor floodplains; these 7 represent relatively long-term sediment storage compartments. Stream channels lead into either larger river systems, lakes, or the Pacific Ocean. Transport of fine sediment through large river systems will not be considered in this overview. Gully channels are the primary conduits of hillslope-derived sediment into the drainage network for the following three reasons. Firstly, total channel length of a given order within a drainage basin typically increases with decreasing channel order; therefore, low-order channels (ie. gully channels) simply have the most length with which to intercept hillslope-delivered sediment (Strahler 1957, Duncan et al. 1987, Bilby et al. 1989). Secondly, higher order streams are often at least partially buffered from hillslopes by toe-slopes and valley- flats. Finally, gully presence is both due to, and enhances, hillslope process activity; therefore, gullies are zones of preferentially concentrated hillslope erosion. 1.3.2 Hillslope to gully-channel sediment transfers The hillslope to gully-channel cascade component consists of processes which deliver sediment from open slopes and gully walls to storage sites immediately accessible by streamflow in the gully channel. Hillslope to gully-channel sediment transfer processes are listed in Table 1.1, based on process lists found in Pacific Northwest sediment budget studies (Reid et al. 1981, Lehre 1982, Roberts and Church 1986, Caine and Swanson 1989, Millard 1993). Also listed are the factors controlling the short-term temporal and spatial occurrence of each process. Temporal occurrence of each process is controlled by the occurrence of specific ranges of hydrometeorological conditions. Spatial occurrence of each process is restricted to sites having certain topographic, vegetative, and soil characteristics. Typical sites of hillslope to gully-channel fine-sediment transfers are illustrated in a sketch of an idealised coastal British Columbia gully, in Figure 1.1. 8 6 Cu S CL -3 2 o 3 . in B * O o 0 1 s S3 o • B . 0 9 5! O 3 oo o s os 3 o 3 Co 5' 3 co 3: E T o' 3 * O X 3 3. o c o 3 ST* o" B * O » B o 3 S3 O 5i o ? 3 * . o a. 3' n 3 cr c Cu 0 0 o CD • 3 C L 3-o 5-s i B cr cn O •z o 3 oo 3 < 3," o cr <"> S3 3 Cu o 3 •a o S3 3 C u cn •a S3 O C o 3 O o C u 3 . 3 O ^ •a E 3 -8 I" ~ ri ss tj " 2 5= ! B" " S3 "1 E •— o C/) o CD CD < CD o 2 o CO C0_ 0 T 3 CD 1 CQ . C _ •<" 3 to <3 3 ' CO CD CD S CD D . 3 " t | CQ O o o 2. co CD CO < CD . 3 " co' 3 ' CO " O 03 CO D —1 03 < CD 3 . CO = 3 " 3 CD CQ CD ? 03 CO O CO 03 o o 3 > 3 3' 03 o <' CD CD r~* 3 " o o o o 5 o o 5 o Q . CD 3 CD 11 z o to _ CD CD :• CD O CD 00 ?! o <ci 3 " CD CD Q) CO JS-— CD Q . £2. o •o CD CO CD 03 T3 C 3 CD 5" Q- 3 < - * 3 CD 03 CD CQ =-' 03 CD § i f 03 CD CO C/J 5-12 "a 2. co 03 " 00 z c CQ 3 3 " < _ _ CD T3 (Q » re 3 £ £ °> 03 Q , E w CD CO < 03 2 . 03° CT CD 3 " r i 3 CO CD C L C0_ 0 T 3 CD 1 CQ c o 03 C L . CO 03 3 o cr fD CD —* CO CO II S S < CD 8 § O CO •-•« o CO — § 3 « 2. O o < CD 03 3 O CO CD o CD CO ' 3 o ? o o < CD 03 zi. o CQ 7T 3 " O CO 03 o CO = : ^ . §3-1 o 55. CD < 03 5' CD CQ C Q ' =• 3 " CD <B — 3 3 . C L 3 CO O -55. c CD CD CQ* O Q . CD S CD O a CD S CD O C L CD —1 03 «-» CD O Q . • CD S CD O 3 O 03 < "2. CD Q_ » I S CO CO CO o ^ — o 03 C L a 5? CD 03 CD CD O •IS » < CD —\ CD ~ ~ co =• 0 s ft) 03 03 CQ 3 " o 3 g co" c —1 CD 3 "-C L < CD —1 O 03 CD o o CD CO i -si 3 - D I CD O O O c CO 8 C L —s O o 3 3 2-cr CD 2. % co o o CO o' 03 CD CD < CD «5 3 -CQ* 3 " o C L CD —1 03 CD O C L CD s CD 3 r -° i C L < CD j j --1 o 03 CD o C L CD —1 03 CD L J < C L < CD cr -1 o 03 CD i f C L < CD cr CD CD i f C L < CD cr —1 o 03 CD* • Q TD S 51 CD cr. 13*1 Headwall Failure Scar Figure 1.1. Idealised coastal British Columbia gully morphology. Typical hillslope to gully-channel fine-sediment transfer sites are shown. 10 1) Rapid mass movements Relatively large, infrequent mass movements are the dominant "episodic" sediment transfer processes on the hillslopes of coastal British Columbia (for example, Rood 1984). These consist primarily of open-slope debris slides and gully-confined debris torrents, the latter often being triggered by gully-wall slides into debris-charged gully channels (Bovis and Dagg 1987). However, smaller, more frequent rapid mass movements which deliver sediment from gully sidewalls and headwalls to storage compartments adjacent to gully channels may be considered to occur in the normal regime. Rapid mass movements can occur anywhere along gully walls, but they occur preferentially in hillslope concavities, often referred to as zero-order basins, hollows or depressions (Dietrich et al. 1987). Sediment gradually accumulates in the gully-wall depressions, followed by periodic sediment discharge to the gully channel below by rapid mass movement. If the accumulated sediment remains stable long enough for vegetation to become established, then stability is enhanced; in this case, accumulation may continue for years or decades, until a large, "episodic" debris slide occurs again. On the other hand, thin veneers of accumulated sediment may frequently slide or flow off former failure planes ~ usually impermeable depression floors ~ never allowing vegetation to become re-established. In such cases, the shallow mass movements can be considered "normal-regime" processes. Temporal occurrence of normal-regime rapid mass movements is primarily controlled by high pore-water pressure in the gully sidewall and headwall soils. 2) Treethrow Treethrow can be a locally significant hillslope to gully-channel sediment transfer process in forested gullies in coastal British Columbia, when trees fall immediately adjacent to gully 11. channels. Biological factors which weaken or kill trees, and timber harvesting activities which alter tree exposure to wind, are the most important controls on treethrow occurrence. Given these independent controls, trees fall preferentially during high winds and when soils are wet (Swanston 1991). 3) Animal activity Animal activity consists of sediment transfers by mammal burrowing (especially by bears and rodents), mammal trampling (especially by ungulates), and invertebrate tunnelling (especially by earthworms). During the study period, I observed black bears (JJrsits americanus) digging near channels in Russell Creek Basin on two occasions. I also observed numerous small herds of Coastal Blacktail deer (Odocoileus hemionus ssp. columbianus) and occasional large herds of Roosevelt elk (Cervus canadensis) in Russell Creek Basin and vicinity. Game trails were evident in both study gullies, but the erosional impact of these appeared to be minor. 4) Frost action Frost action is the loosening and detachment of soil material by the growth of needle ice, followed by the downslope movement of the detached material after needle ice thaw. Downslope movement may result from the following (Mackay and Mathews 1974, Lawler 1993): 1) differential ice uplift and gravitational fall of soil particles (perpendicular to slope and closer to vertically down, respectively), 2) downslope toppling of ice needles as they melt, 3) downslope sliding of ice needles and incorporated sediment during thaw, 4) sediment transport by sediment-laden rivulets of needle ice meltwater, 12 5) transport by other processes (eg. rainsplash, sheetwash, dry ravel) of loosened soil "skin" left in situ on frost-loosened slope. Frost action requires high soil moisture content, and cycles of sub-freezing and above-freezing air temperatures, to allow needle ice growth and ablation. In addition, the soil must be porous enough to permit sufficiently fast migration of soil moisture to the freezing front to prevent simple in situ freezing of soil water in pores (Outcalt 1971). Studies indicate that frost action is not maximized simply by number of freeze-thaw cycles; rather, prolonged spells of very low temperatures are required (Selby 1985). Coastal British Columbia experiences relatively mild winters, with air temperatures frequently near 0°C at low elevations; however, colder spells, several days in duration, occur periodically when cold continental airmasses break out to the coast. Needle ice has been reported in the literature in coastal British Columbia at elevations from near sea level up to alpine (Lawler 1988). Frost action is an important hillslope to gully-channel transfer process on unvegetated slopes, where ice-loosened sediment can travel unimpeded downslope following needle ice ablation. Frost action occurs at sites such as active gully sidewalls, road cutbanks and road fillslopes, where these are composed of sufficiently porous (sandy-matrix) soils. Frost action is probably more common in clearcut areas than in forests because of the greater nocturnal radiative cooling in clearcuts. 5) Sheetwash and rilling Sheetwash is the entrainment and transport of sediment down exposed soil slopes by overland flow. Planar surfaces subject to sheetwash generally develop small preferential water 13 and sediment transport pathways called rills, which concentrate the erosive power of the overland flow, thereby enhancing their own growth (Selby 1985). Sheetwash and rilling are temporally controlled by the occurrence of overland flow; therefore, antecedent soil moisture and rainfall intensity and duration are important controls. The presence of snow-cover can presumably have ambiguous effects on the occurrence of overland flow. Snow cover may absorb much of a short, intense rainfall burst, decreasing the rate of water reaching the soil surface, thereby decreasing the rate of overland flow; on the other hand, snow-melt during a prolonged intense rainfall may provide additional water to the soil surface, increasing the rate of overland flow. If large volumes of water reach the soil surface, blockage of overland flow by the snowpack is probably not a factor, as the water can thermally erode flow courses. Overland flow and sheetwash may also occur when rain falls on frozen soils. Numerous authors have noted the almost total absence of overland flow in undisturbed coastal British Columbia forests, because infiltration rates usually greatly exceed rainfall intensities, and because small ground surface hollows tend to cause surface water ponding when intensities do occasionally exceed infiltration rates (Cheng et al. 1975). Only relatively impermeable surfaces generate overland flow, and many of these — such as bedrock outcrops — lack an erodible soil cover. The most common sheetwash / rilling sources are compacted, unpaved logging road surfaces and failure scar surfaces in compact materials such as till (Swanson et al. 1987). Fine-sediment transfers from road surfaces by sheetwash are largely controlled by the level of road activity, with active roads being the greatest contributors (Reid and Dunne 1984). However, overland flow generated on road surfaces may also entrain sediment from road fillslopes where road-surface runoff is diverted downslope. Rilling of unconsolidated road fillslopes is common in coastal British Columbia. Failure scar surfaces in coastal British Columbia are typically composed of impermeable glacial till; their low infiltration rates are what caused high 14 pore water pressures and subsequent failures to occur there. Even if the till failure scar surface is resistant to sheetwash erosion, the overland flow generated there may entrain sediment veneer material that has accumulated on the scar. 6) Dry ravel Dry ravel is the downslope, gravitational saltation of dry sediment clods which become detached from an unvegetated soil slope. Dry ravel probably occurs mainly on sidewall and headwall failure scars, active sidewall slopes, road cutbanks and road fillslopes (for example, Reid 1981). Although dry ravel is often included as a form of soil creep (Dietrich et al. 1982, Roberts 1984), separate mention dry ravel seems appropriate because of its more concentrated spatial occurrence. 7) Rainsplash Rainsplash is the propulsion of sediment particles into the air, in all directions, when raindrops hit an exposed soil slope, and the subsequent net downslope movement of particles as they return vertically to the slope surface. Rainsplash also indirectly increases the efficiency of sheetwash transport by increasing the turbulence of overland flow (Selby 1985). Rainsplash is temporally controlled by the occurrence of intense rainfall to splash sediment particles, and by the absence of snow cover sheltering soil surfaces from raindrop impact. Cohesive soils are more resistant to rainsplash when they are already wet before intense rainfall (Olive and Rieger 1985); however, antecedent soil moisture is probably less important in coarser-grained soils. Rainsplash occurs on unvegetated soil slopes. Vegetation shelters the soil surface from raindrop impact and interferes with the downslope progress of splashed particles. Rainsplash is 15 most effective on non-cohesive soils surfaces, especially if antecedent soil moisture is high. Rainsplash sediment-transfer sites include active sidewall slopes, sediment veneers on failure scars, road cutbanks, and road fillslopes. 8) Soil creep Soil creep is the very slow, widespread, downslope movement of the upper layer of the hillslope soil mantle due to several processes, including gravitational shear stress, expansion/contraction cycles due to wetting/drying and freezing/thawing, excavation and infilling of root holes, and the transfer of wind stress to the soil via tree roots (Carson and Kirkby 1972). Temporal patterns in soil creep activity are controlled primarily by temporal hydrometeorological cycles of soil wetting/drying and freezing/thawing (Barr and Swanston 1970, Carson and Kirkby 1972, Anderson and Cox 1978). The biogenic processes are generally less significant than the hydrometeorologically controlled expansion/contraction cycles, and gravitational shear stress is constant at the time scale of interest. Swanston (1981) reported maximum soil creep activity in the rainy autumn-winter season in the Coast and Cascade Ranges of Washington, Oregon, and northern California. Presumably, soil creep in coastal British Columbia is also greatest during the wet winter season when soils are moist and freeze/thaw cycles are common. Sediment transfer from hillslopes into gully channels due to soil creep probably occurs along almost all gully-channel margins, wherever a soil mantle is present on the gully walls. Soil creep occurs primarily between clay particles (Carson and Kirkby 1972); therefore, soil creep rates may vary locally depending on clay abundance (Swanston 1981). Clearcutting may increase soil creep rates due to loss of root strength and alteration of hydrometeorological conditions (Wu et al. 1979, Swanston 1981). 16 1.3.3 Fluvial sediment transport in gully channels Fine sediment may be transported down gully channels, from channel-margin or in-channel storage sites to streams, by two processes: fluvial suspended sediment transport or debris-torrent transport. Both processes involve temporal cycles of sediment accumulation and transport which occur at different frequencies. Fluvial suspended sediment transport, occurring much more frequently than debris-torrent transport, is considered a normal-regime sediment transfer process, while debris torrents are usually considered as individual episodic events. Comparison of fluvial and debris-torrent sediment transport scales "Debris torrent" is a term which has been used in western North America in the last 20 years to refer to highly fluidised, gully-confined debris flows with significant LOD (large organic debris) content and low proportion of fine sediment (Swanston and Swanson 1976, Hungr et al. 1984, VanDine 1985). Debris-torrent activity cycles consist of relatively long sediment-recharge periods, during which sediment accumulates in gullies, especially behind L O D pieces or jams, followed by episodic flushing of the accumulated sediment by debris-torrent transport (Oden 1994). During the recharge period, fluvial sediment transport operates at much higher frequency related to runoff generation, preferentially removing fine sediment from gully storage compartments, leaving coarsened deposits to be removed by debris torrenting. Comparisons of fluvial to debris-torrent transport need to consider the scale differences of the four dimensions discussed in Section 1.2: space, time, magnitude and sediment texture. Fluvial sediment transport is spatially and temporally more frequent than debris-torrent transport. Fluvial transport occurs (or occurred in the past) in all gullies, while only gullies meeting certain sediment-input, morphologic and hydrometeorological characteristics are prone to debris-17 torrenting. On the other hand, importance of debris-torrent transport derives from the much greater magnitudes involved in individual events. However, debris torrents consist of a much greater range of sediment textures, more dominated by the coarse fraction, than fluvial transport. In a coastal British Columbia example, Millard (1993) compared fluvial and debris-torrent total-sediment transport in two Coquitlam Valley gullies. Average transport by individual debris torrent events equalled 600 years of fluvial transport, presumed to be much longer than average torrent recurrence interval. However, the results would probably have been much different if only fine sediment had been considered. Millard found that torrent-prone gullies in Coquitlam Valley had much greater hillslope-derived sediment inputs and fluvial fine-sediment outputs than stable gullies. Fluvial coarse-sediment outputs from the two gully types, however, were almost equal. Therefore, the hillslope-derived coarse-sediment inputs in the torrent-prone gullies were almost entirely being stored within the gullies. Thus, during the ensuing debris torrent, stored material contribution to torrent output would have been almost entirely coarse. However, sidewall slides triggering debris torrents represent unsorted sediment inputs; the fine fraction of these may be the dominant source of fines in debris torrents. Summarising, debris-torrent transport may be much more important than fluvial transport in torrent-prone gullies, when total sediment outputs or coarse-sediment outputs are considered. However, the difference may be less, or indeed may be reversed, when only fine sediment is considered, especially if sidewall failures triggering torrents represent small fractions of total torrent volume: Spatial and temporal controls on suspended sediment transport in gullies Considering only the normal regime now, occurrence of suspended sediment transport in gullies is spatially controlled by the distribution of channel-margin and in-ehannel sediment 18 storage compartments. Temporally, suspended sediment transport depends on both flow conditions and sediment-supply conditions. Channel-margin sediment storage compartments typically take the form of colluvial cones, fans and aprons at the bases o f active sidewall slopes and failure scars. Spatial occurrence o f fine-sediment storage sources within gully-channels is largely controlled by L O D jams in coastal British Columbia (Oden 1994). Typically, one or a few large "key" pieces o f L O D become stuck in a gully-channel and trap smaller pieces that are carried down-channel, until a dense jam has formed. Then fluvially transported sediment accumulates behind the jam, until a "wedge" o f sediment has formed. L O D jams and associated sediment wedges vary greatly in scale, from a few tree branches retaining a few cubic decimetres of sediment, to several large tree trunks retaining hundreds o f cubic metres o f sediment. Supply of gully channel-margin sediments varies in time, controlled by the various hillslope to gully-channel input processes and by the removal of stored sediment by previous gully-channel runoff events. Gully-channel runoff events caused by rainfall, rain-on-snow or radiation snowmelt transport the majority of suspended sediment for two reasons. Firstly, high flows are more energetic than low flows, thus they have greater capacity to transport sediment in suspension. Secondly, high flows (or, more precisely, high stages) access channel-margin sediment supply which has not been exposed to entrainment by previous lower flows. The latter point may be more important because gully channel-margin sediment supply is not necessarily replenished between each high flow event in a gully channel. Since upland channels normally carry far less than full capacity suspended sediment loads, access to sediment supply is probably the primary factor controlling transport (Van Sickle and Beschta 1983). In-channel sediment storage sites — such as L O D sediment wedges — are depositional features created during low to moderate gully channel flows. Sediment is entrained and 19 transported down-gully by flows which are sufficiently high to entrain the sediments from the low-gradient wedge surface. Less frequently, very high flows may disrupt the structure of the LOD jam, suddenly permitting rapid erosion of the wedge sediment and resulting in large down-gully transport loads. 1.3.4 Fluvial sediment transport in streams Fine sediment transport in main stream channels occurs almost entirely as fluvial suspension, although debris torrents may travel all the way down gullies and along main stream channels in some circumstances. As in gully channels, fluvial suspended sediment transport is considered to be the normal-regime transport process. The normal regime is interrupted less frequently by episodic events in streams than in gullies because of the buffering effect of increasing drainage area and because of the physical separation of stream channels from hillslopes by toe-slopes and valley-flats. Streams transport suspended sediment delivered by tributary gullies and streams as well as sediment entrained from in-channel storage sites and streambanks. Spatial controls on stream-channel sediment storage sites Fine sediment may be stored in the interstices of gravel deposits in stream channels and is available to be mobilised when subject to flow conditions sufficiently energetic to disturb the gravel. In alluvial gravel-bed stream channels, gravel bars tend to occur at regular intervals on alternating sides of the channel. In coastal British Columbia, however, most streams are not entirely alluvial; rather, their morphologies are controlled by non-alluvial features, such as bedrock knickpoints, large alluvial or colluvial tributary fans, and LOD. These features largely control the location of gravel deposits within stream channels. 20 Spatial controls on streambank erosion In alluvial stream channels, bank erosion is greatest on the outside edges of meander bends. In the largely non-alluvial stream channels of coastal British Columbia, however, sites of bank erosion are typically controlled by impingement of channels against erodible valley walls, particularly when flow is deflected around LOD accumulations. Temporal controls on suspended sediment transport in stream channels Suspended sediment transport in streams is temporally controlled by the input of suspended sediment from tributary gullies and streams, the entrainment of fine sediment from in-channel bed material deposits, and the entrainment of bank material. Fine sediment in bed material bars is entrained in two phases (Jackson and Beschta 1982). "Phase 1" occurs under low to moderate flows, as fine sediment on the streambed surface is winnowed from amongst the larger, stable particles forming the armour layer. "Phase 2" occurs under high flows, when the armour layer particles are entrained as bedload, exposing all interstitial fine material to suspension. Phase 2 bedload has been observed to require greater shear stress (directly related to discharge) to be initiated than to be sustained. Therefore, bedload generally begins at relatively high flow on the rising limb of an event hydrograph, and does not cease until a much lower discharge on the falling limb (Reid and Frostick 1984). Timing of interstitial sediment exposure to flow is also controlled by periodic disruptions of LOD jams, which generally occur during high flow, but are episodically occurring events not directly related to discharge. When LOD jams break or shift, previously buttressed sediment suddenly becomes available for transport (Mosley 1981, Heede 1985, Sidle 1988, Smith et al 1993). Bank material entrainment is related temporally to the occurrence of high flows capable of attacking rarely accessed material and transporting large quantities of the material downstream. 21 Antecedent hillslope processes such as frost action and dry ravel, as well as antecedent removal of the resulting detritus, are also important in the bank erosion process. 1.4 Short-term temporal patterns in fine-sediment transfers 1.4.1 Hydrometeorological conditions Hydrometeorological conditions provide, directly or indirectly, the main temporal control on almost all normal-regime fine-sediment transfer processes; therefore, temporal patterns in hydrometeorological conditions tend to be reflected in sediment transfers. Short-term, temporal patterns in hydrometeorological conditions consist of cycles operating at seasonal, sub-seasonal, and synoptic scales. The climate of coastal British Columbia is defined by two seasons: moderately warm, dry summers and cool, wet winters (Oke and Hay 1994). The two seasons have characteristic, dominant synoptic-scale weather systems (Suckling 1977). Synoptic scale weather systems are defined as having typical time scales of one to several days and typical spatial scales of hundreds of kilometres (Henderson-Sellers and Robinson 1986). In the coastal British Columbia wet season, typical synoptic-scale weather systems consist of: 1) frequent eastward-tracking cyclonic storms from the Pacific, resulting in moderate to heavy rain and/or snowfall (typical freezing level at 300 - 800 m) — Suckling's "ocean lows"; 2) inter-cyclonic lulls dominated by Pacific airmasses, resulting in mild, dry to moderately wet, weather - Suckling's "unclassified systems"; 22 3) occasional southwestward extension of the continental polar (or continental arctic) anticyclone to the coast (cP / cA "outbreak"), resulting in cold, dry weather --Suckling's wintertime "land highs"; 4) occasional northward extension of the subtropical anticyclone, resulting in mild, dry weather -- Suckling's "ocean highs". In the coastal British Columbia dry season, the subtropical anticyclone becomes far more dominant, Pacific cyclones become far less frequent, and the cold continental anticyclones do not occur (Suckling 1977). The sub-seasonal scale represents cycles of synoptic-scale hydrometeorological conditions which appear in the coastal British Columbia wet season. The cycles consist of periods of cyclonic storms and inter-cyclonic lulls (multiple synoptic-scale events), alternating with cP airmass outbreaks (single synoptic-scale events). The punctuation of the more frequent cyclonic / inter-cyclonic sequence by the more infrequent cold, dry spells is important for two reasons. Firstly, air temperature remains near or below freezing for one to several days during cP airmass outbreaks. Secondly, cyclonically generated precipitation at low elevations (less than 800 m) often occurs in the form of snow immediately preceding and following cP airmass outbreaks. Both of these conditions are otherwise unusual occurrences in coastal British Columbia. Thus, snowpack accumulation/ablation cycles are closely associated with the sub-seasonal temporal scale at low elevations in coastal British Columbia. 23 1.4.2 Hillslopes to gully channels Seasonal and sub-seasonal scales At the seasonal scale, hillslope to gully-channel fine-sediment transfer behaves differently between summer (dry season) and winter (wet season). Of the processes described in Section 1.3.2, summertime conditions favour only dry ravel. Although other processes may occur in summer, their occurrence is favoured by wintertime conditions. Within the wet winter season, sub-seasonal hydrometeorological patterns control hillslope to gully-channel, fine-sediment transfer processes. Rapid mass movements, rainsplash, sheetwash, and treethrow occur preferentially during cyclonic rainstorms, due to intense rainfall, overland flow, high soil moisture and high winds. Cold cP airmass outbreaks between rainy periods favour the occurrence of frost action, which loosens soil surfaces, preparing them for erosion. Therefore, cyclonic rainstorms following these cold spells are especially favourable for the occurrence of rainsplash and sheetwash and mass movements. The occurrence of snowfall associated with the cold spells is also important. Snow accumulation at the beginning of a cold snap will inhibit frost action on soil surfaces which are insulated from the cold air by snow cover. On the other hand, the first rainstorm following the cold snap will result in additional water delivery to the ground surface if snow is present, thereby enhancing overland flow, soil moisture and the sediment transfer processes which these conditions favour. Snow accumulation at the end of a cold spell may also have similar results, but without having inhibited frost action during the cold spell. A final case exists whereby snow cover is present at the beginning of a cold spell, except on steeply sloping surfaces, such as active sidewall slopes and road cutbanks. Then frost action may occur during the cold spell, leading to sediment 24 transfer toward the gully-channel after thawing; however, snow cover at the bases of sediment sources intercepts the downslope transfer of fine sediment. Only when the snow has completely melted is the sediment transfer completed. Table 1.2 provides examples from the literature of seasonal and sub-seasonal patterns in fine sediment transfers from hillslopes to gully channels and small stream channels. In most of the Pacific Northwest examples, summer is assumed or observed to be the period of dominant sediment accumulation at channel margins. This is not necessarily the period of greatest sediment transfer activity, however, as wintertime entrainment of channel-margin sediment may obscure much greater hillslope to channel delivery rates during this period. The observation of seasonal depletion of channel sediment loads during the winter runoff season, however, indicates that summertime storage must be of considerable magnitude. However, none of the Pacific Northwest examples is based upon measured data. Synoptic scale The behaviour of hillslope to gully-channel sediment transfer processes during synoptic-scale events is poorly understood. This arises largely from the impracticality of measurement. Typical methods of measuring hillslope-derived transfers involve repeated surface surveys or sediment catchment arrangements. These are relatively crude measures when analysed at very short time scales. Efforts have been made to electronically record erosion of hillslope sources by means such as photoelectric erosion pins (Lawler 1992), but these are not yet widely employed. 1.4.3 Gullies and streams to basin outlet Temporal patterns in suspended sediment transport in gully and stream channels are controlled by temporal patterns in hydrometeorological conditions in two ways: 1) temporal 25 Table 1.2. Literature summary: Seasonal and sub-seasonal patterns of fine-sediment transfers in small, normal-regime catchments Site Hillslope-to-Channel Transfers Through-Channel Transport Region Dr. Area (ha) Terrain Description Sediment Sources (1) Season Process Season Process Seasonal Depletion Reference Pacific Northwest S.E. Alaska 154 Mtnous, stable, forested Channel only — — Autumn Frontal rain & rain-on-snow None Sidle and Campbell, 1985 Vane. Isl., B.C. 1020 Mtnous, unstable, partly logged Failure scars, roads, gully sidewalls Summer? ? Autumn -winter Frontal rain & rain-on-snow Strong (2) Tassone, 1987 W. Oregon 75 Mtnous, stable, partly logged Burnt slopes, roads Summer Winter Dry ravel Sheetwash Autumn -winter Frontal rain & rain-on-snow Moderate Beschta, 1978 W. Oregon 202 Mtnous, stable, forested Channel only Summer? ? Autumn -winter Frontal rain & rain-on-snow Strong (2) Beschta, 1978 W. Oregon 750 Mtnous, unstable, forested Failure scars, gully sidewalls Summer Dry ravel Autumn -winter Frontal rain & rain-on-snow Strong Paustian and Beschta, 1979 Rockv Mountains W. Montana 68 Mtnous, stable, forested Channel only -- — Spring Rad. snowmelt, frontal rain-on-snow Weak Anderson and Potts,.,1987 W. Montana 137 Mtnous, stable, partly logged Road sidecast Summer -autumn Dry ravel Spring Rad. snowmelt, frontal rain-on-snow Strong Anderson and Potts, 1987 W. Alberta 1580 Mtnous, alpine-subalpine Active valley sidewalls Early spring Sheetwash Frost action? Late spring Rad. snowmelt Strong Nanson, 1974 Basin and Ranae N. Arizona 186 Hilly, scrubby forest Hillslopes Autumn -winter Winter Sheetwash Frost action Autumn -winter -spring Frontal rain & rain-on-snow, rad. snowmelt Strong / subseas.? (3) Lopes and Ffolliott, 1993 Midwest U.S.A. Iowa 30-60 Rolling plains, cultivated Gully sidewalls Spring Winter Sidewall slumping Frost action Spring -summer Convective rain Strong Piestetal, 1975 Indiana 0.05 Gullied mine spoils Gully sidewalls Winter Frost action, wetting-drying Spring -summer Convective rain Strong Olyphant et al, 1991 Europe ( S. England 0.01 -0.10 Disturbed heathland Gully sidewalls Winter Summer Frost action Dry ravel, wildlife Winter Summer Frontal rain Occ. frontal rain Subseas. (3) Tuckfield.1964 N. Italy 75-1000 Hilly, cultivated vineyards & grains Gully sidewalls, plowed land Winter Summer Frost action Field sheetwash Winter -spring Summer Frontal rain-on-snow Convective rain Subseas. (3) Tropeano,1991 Australia New S. Wales 76 - 225 Hilly, forested Gully sidewalls Yr. - round Wetting - drying, then sheetwash Summer Winter Convective rain Frontal rain Subseas. (3) Olive and Rieger,1985 cn^ 430.wb1 Notes: 1. Sediment sources : a. "Channel" incl. bed, banks, and channel storage (incl. LOD wedges in Pac. N.W. and Rocky Mtns.) b. Assume all channels have "channel sources", only listed if no other sources. c. "Roads" incl. surfaces, cut-banks, and sidecast fillslopes. 2. Occurrence of successive max.-to-date runoff events controls strong seasonal depletion. 3. Subseasonal depletion occurs, ie. sediment depletion/replenishment cycles occur at higher-than-seasonal frequency. control of transfers from hillslopes to channel-margin storage compartments, and 2) temporal control of gully-channel runoff. Seasonal and sub-seasonal scales Small-channel suspended sediment transport occurs almost entirely during high runoff seasons (ie. during the wet winter season in coastal British Columbia), due to the importance of high flows in transporting sediment. Within high runoff seasons, suspended sediment typically shows a pattern of depletion over the course of the season, or a sub-seasonal cycle of depletion and replenishment. Season-long depletion is due to delivery of fine sediment to channel-margin storage compartments during the low flow season, followed by the flushing of the stored sediment down-channel by early-season high flow events. Later-season high flow events have little channel-margin sediment available to transport; therefore, suspended sediment concentration for given discharge declines as the runoff season progresses. Sub-seasonal patterns are due to replenishment of channel-margin sediment storage by hillslope processes throughout the runoff season. The supply usually does not meet removal rate, so that depletion by successive runoff events does occur; however, certain hydrometeorological conditions during the runoff season favour the replenishment of channel-margin sediment reserves. Table 1.2 provides examples of seasonal and sub-seasonal patterns of suspended sediment output from small catchments, due to the interaction of hillslope to channel transfers and channel runoff. Dominant seasons of sediment transfers and flushing vary from region to region, but that each region exhibits a specific pattern. 27 The coastal British Columbia seasonal and sub-seasonal scale pattern is represented in Table 1.2 by the Pacific Northwest regional examples. The five regional examples list autumn -winter as the runoff season, in channels draining catchments of 75 to 1020 hectares. In four of the studies, moderate to strong depletion of suspended sediment transport occurred over the course of the season. In these channels, sediment storage accumulation at channel margins must have occurred dominantly during low flow season (spring - summer); relatively little recharge of channel-margin storage can have occurred during the runoff season, otherwise continued seasonal depletion would not have been evident. Dry ravel is the only hillslope to channel sediment-transfer process which occurs preferentially during the dry season (refer to previous section). The majority of the hillslope to channel sediment-transfer processes occur either in conjunction with runoff-generating hydrometeorological conditions (e.g., mass movements, bank erosion, sheetwash, during intense rain and/or snowmelt), or in between runoff-generating conditions (eg: frost action during cold snaps between stormy periods). Therefore, in the four regional examples cited, in which suspended sediment transport diminished during the runoff season, hillslope to channel sediment-transfer processes were probably dominated by summertime dry ravel and/or autumn rainsplash on dessicated soils. In the fifth example (Bambi Creek, Alaska), in which seasonal depletion of sediment storage did not occur, external sources of sediment were negligible; therefore, sediment transport was probably controlled almost directly by the ability of stream runoff to erode streambahks and disrupt LOD jams (factors more directly related to channel discharge). The other examples were included in Table 1.2 to illustrate the regionally specific nature of temporal relations between hydrometeorological conditions, hillslope to channel fine-sediment transfers, and along-channel fine-sediment transport. The European examples are of specific interest, however, because the sub-seasonal patterns exhibited in those studies could presumably 28 occur in coastal British Columbia. In both cases, frost action during cold snaps, within the winter runoff seasons, was identified as the most important contributor of fine sediment to channel margins. As a result, steady seasonal depletion of channel margin sediment storage did not occur during the runoff season; rather, sediment was removed by rainfall and/or snowmelt generated runoff, and occasionally replenished by frost action several times over the course of a runoff season. Documented examples of this pattern in coastal British Columbia have not been presented in the literature. Synoptic scale Most suspended sediment transport in small channels occurs during discrete runoff events (synoptic scale). Synoptic-scale temporal patterns of suspended sediment transport within these runoff events usually consist of sporadic transport bursts (Bley and Schmidt 1991), the patterns of which may be controlled by antecedent hydrometeorological conditions. Patterns at the synoptic scale are typically expressed in terms of chronological discharge (Q) versus suspended sediment concentration (SSC) relations, essentially comparisons of Q : SSC ratios on rising versus falling limbs of individual runoff event hydrographs. Table 1.3 presents a compilation of Q vs. SSC relations in channels draining small catchments discussed in the literature. The most common Q vs. SSC relation cited in the literature, by far, is the single or multiple clockwise hysteresis loop. In events with this hysteresis loop pattern, the majority of suspended sediment transport occurs during the rising limb of the event hydrograph, as opposed to the falling limb. Clockwise hysteresis can be attributed to sediment supply depletion during the course of the runoff event. While the absolute amount of sediment available in channel margin storage appears to depend on seasonal and/or sub-seasonal trends of accumulation and depletion, a further pattern of sediment accumulation between runoff events, and depletion during runoff 29 s cn CO * i $ I § c T3 Tl •< I . 3' e. D 9° co co o 3 < « a 8 & 3 8 c 3 <D 8 i * CD g- § a c 3. 3 ID re < re 3 D I I P- 3' e o 5- ? to »• re <* =i C/) a a o ro < o 3 8. •S £ o XI 3 "8 I o- o 9 8 c5 X) o c T) TO *C a c a re •o ro cr. o 3 o a ro re K. ~ 8 •a a ro 1? a ro 3 o 3 C CO 5' CQ ro o c 2 ro c 3 1 f z CD CO I ro 3 i CO I o CD 3 CD CD ro 3 re s a a co g § < re g> | a g. b, 3 8 o cn o ° cn ro 0) C 3 8 3 3 o 3 9 3 3 o 3 9 3 3 o 3 ? 3 3 o 3 ft? c 3 8 3 3 o 3 0 ? re 1 co 00 cn CD oo CO i' rp ro' CD ro co ca cn o 8 * re CO CO 03 (0 i - :p f?_ 111 CO CO 03 ~g cn oo CO Tl > Tl I i n 5 » re JL 3" 8 3 ro fl» CO oo cn events, appears to be very common. The severity of sediment supply depletion may be judged by the relative width of the Q vs. SSC hysteresis loop. This pattern may be enhanced by the vertical particle size distributions in storage compartments; in storage units created by gravitational deposition (eg: channel-margin talus cones and aprons), texture is likely to coarsen downslope; therefore, as rising flows access previously untouched sediment within the storage compartment, they are also accessing finer material which is more easily mobilized. Counter-clockwise hysteresis loops occur when sediment transport is somehow delayed with respect to the timing of discharge. In small drainage basins, where precipitation is essentially uniformly distributed in time over the basin and travel time differences for flood waves and sediment are negligible, sediment transport delay is most likely linked to some requisite threshold for sediment entrainment. Two possible examples include sediment supply protected by, or held within, the snowpack requiring snowmelt before sediment becomes available for transport; and fine sediment protected by a coarse armour layer or LOD jam which breaks up near peak flow, allowing fine sediment to be carried into suspension at the peak and during diminishing flows. Figure-8 hysteresis loops can result from sediment supply depletion early in a runoff event, followed by replenished supply later in the event. For example, channel-margin storage compartments may be depleted early in the event by rising channel discharge but then, near peak flow, LOD jam failure and/or gully sidewall failures provide fine sediment directly to the channel. In any case, two modes of sediment transfer to the channel are required. 1.5 Land management and temporal patterns in the fine-sediment cascade In coastal British Columbia human activities associated with natural resource extraction have altered the natural sediment cascade. In particular, much attention has focussed on the effects that forestry-related sediment-cascade alterations have had on salmonid habitat in gravel-31 bed streams. While other issues are locally important, the fish-forestry issue applies to most of coastal British Columbia and has been chosen as ah issue to which the results of this study can be applied. 1.5.1 Forestry and fine sediment Forestry activities can alter the fine-sediment cascade by increasing the incidence of landslides (episodic events), as well as by creating more sediment sources -- including landslide scars and road networks ~ from which fine sediment can be chronically transferred to the drainage network under normal-regime conditions. Forestry also affects the fine-sediment cascade by altering hydrologic patterns and LOD regimes in channels. Numerous studies have reported increased incidence of landslides after logging, and especially after logging-road construction, in the Pacific Northwest and in other humid-temperate, mountainous regions (for example: Swanson and Dyrness 1975, Reid et al. 1981, Rood 1984, several in a review by Sidle et al. 1985). Reid et al. (1981) noted that landslides were the dominant mode of coarse-sediment transfer from hillslopes to channels in two Olympic Peninsula, Washington, study watersheds; however, when considering only fine-sediment transfers, secondary erosion of landslide scars accounted for about one-third the quantity of erosion that occurred in the original landslide events. Swanson et al. (1987) also noted the importance of landslide scars in producing fine-sediment during normal-regime. Besides being the main cause of logging-related landslides, logging roads themselves are usually the greatest logging-related contributors of normal-regime fine sediment to drainage networks (Reid 1981). Studies of forestry-related impacts on hydrology in the Pacific Northwest have generally shown increased peak flows and decreased lag times to peak flow during runoff events (Hetherington 1982, Megahan 1983, Golding 1987, Wright et al. 1990). These results were 32 attributed to decreased forest-canopy interception, increased snowpack accumulation, increased snowmelt rate, increased overland flow on road surfaces, interception of sub-surface flow by road cut-banks, and drainage-network expansion in the form of roadside ditches. However, in some cases, peak flows were found to decrease after logging, possibly due to the disturbance of sub-surface macropore drainage by compaction (Golding 1987, Wright et al. 1990) and/or increased runoff storage in logging-related surface depressions, such as low-gradient ditches and rutted machine trails (Church, pers. comm.). Assuming that increased peak flows after logging are the norm, the following implications for fine-sediment transfers through the sediment cascade can be expected: 1) channel-margin sediment storage units can be expected to be accessed by peak flows more frequently, 2) LOD disruptions and sediment-wedge scouring can be expected to occur more frequently, and 3) fine-sediment transferred into channels can be expected to travel further before deposition than previously would have occurred. Forestry affects the storage and transport of fine sediment in channels by altering LOD regimes. Likens and Bilby (1982) proposed a model of the temporal post-logging patterns of LOD loading in channels of various sizes which have been logged to the banks. In small channels ~ including gully channels ~ a loading peak occurs shortly after logging due to input and accumulation of logging slash (broken branches, small broken stems, bark chunks, etc.). Slash typically consists of smaller pieces than are delivered from undisturbed forests. The smaller pieces are more mobile and tend to accumulate in larger jams (Hogan 1987). After several years to a few decades, a second LOD loading peak occurs due to the input of dead primary-stage tree stems. Over the long-term, L O D loading drops to the pre-logging level. In intermediate-sized streams, LOD loading declines following logging because the forest supply has been removed, and the slash pieces are too small to accumulate in the channel. However, the intermediate channel experiences the loading peak associated with the primary-stage stem input. Over the long-term, 33 LOD loading drops to the pre-logging. level. In large streams, LOD loading declines after logging and does not recover until climax forest has regenerated many decades later. Slash and primary-stage stem loading peaks do not occur because the pieces are to small to accumulate in large channels. 1.5.2 Fine sediment and salmonid habitat Salmonids have specific requirements in gravel-bed streams, notably for spawning, shelter, and food production. Increased fine sediment in streams reduces the ability of stream habitat to meet the fishes' needs. While fine sediment is being transported in suspension, stream turbidity is increased. Possible effects of high turbidity on salmonids include difficulty seeing food, reduced food production, and disrupted migration patterns. Bjornn and Reiser (1991) reported that juvenile salmonids will avoid migrating into, and have difficulty feeding in, streams with turbidities as low as 60 to 70 NTU. However, they also concluded that forestry-related turbidity increases are rarely sustained long enough to adversely impact salmonids. Suspended sediment in sufficiently high concentrations ~ on the order of thousands of mg/1 ~ can physiologically damage salmonids, especially their skin and gills (Bjornn and Reiser 1991, Hicks et al 1991). Physiological damage due to high suspended sediment concentrations in streams is rare, occurring only during episodic events, such as immediately downstream of landslides. Deposition of fine sediment into the interstices of gravel bed-material reduces intragravel pore flow of water with several harmful results: fish eggs may suffocate due to insufficient dissolved oxygen delivery, the eggs can be poisoned by their own trapped metabolites, and finally the emerging fry may be physically entombed and eventually starve. Fine-sediment infiltration 34 also robs invertebrates of the intragrayel habitat they require, thus reducing an essential source of food for fish (reviews in Gibbons and Salo 1973, Meehan and Swanston 1977, Tripp and Poulin 1986, Everest et al 1987, Scrivener and Brownlee 1989, and Hicks et al 1991). 1.5.3 Significance of temporal patterns in fine-sediment transfers Short-term temporal patterns of fine sediment tranport through sediment cascades are important because they create temporal patterns in habitat quality and because salmonid habitat requirements themselves vary in time. Short-term temporal patterns in fine-sediment transfer through drainage networks Timing of fine-sediment transfer from hillslopes to channels determines the short-term fate of fine sediment within the drainage network. Timing of fine-sediment transfer from hillslopes to low-order channels controls the immediate downstream travel distance of transferred sediment. For example, Duncan et al (1987) and Bilby et al (1989) demonstrated that the travel distance of road-generated fine sediment through the drainage network was directly related to low-order channel discharge. Eventually, fine sediment that has been transferred to the drainage network will be deposited. When fine-sediment deposition occurs in clast-supported gravel-bed stream substrates, two processes occur (Beschta and Jackson 1979). First, fine particles are deposited in surface-layer (armour-layer) voids between coarse gravel particles. Then, the deposited fines settle downward under the influence of gravity through the smaller pores of the sub-armour substrate. Although fine sediment will be deposited on gravel streambeds more readily during low streamflows, settling through the substrate pores occurs more readily at higher flows, when gravel 35 particles are vibrating. Therefore, maximum fine-sediment infiltration occurs at flows that are not quite competent to entrain the armour-layer gravel particles (Frostick et al 1984). Temporal patterns in salmonid habitat requirements Salmonid life cycles in streams include periods of migration, spawning, egg incubation, and juvenile rearing (Bjornn and Reiser 1991). Timing of salmonid life cycles is species dependent. During migration and rearing periods, salmonids are most vulnerable to high suspended sediment concentrations and turbidities; on the other hand, during spawning and incubation periods, fine-sediment deposition and infiltration in stream substrate is the greatest sedimentological hazard to salmonids. 1.6 Study hypotheses 1.6.1 General hypothesis The general hypothesis of this study is that short-term temporal patterns in fine-sediment transfers through the hillslope-gully-stream cascade pathway in Russell Greek Basin, Vancouver Island, are directly linked to synoptic, sub-seasonal and seasonal scale temporal patterns in hydrometeorological conditions. Given this, the timing* of fine-sediment movement through the sediment cascade of a small coastal British Columbia drainage basin in normal-regime can be qualitatively modelled with reference to hydrometeorological conditions, and local forecasts of relative fine-sediment transfer activity can be made based upon regional hydrometeorological forecasts. 36 1.6.2 Specific hypotheses Hillslope to gully-channel transfers 1) At hillslope sediment sources which consist of non-cohesive banks or slopes, fine sediment is transferred to channels primarily during the dry season by dry ravel, during early wet-season rainstorms by rainsplash, and during winter cold snaps by frost action. 2) At hillslope sediment sources which consist of thin sediment veneers over impermeble surfaces, fine sediment is transferred to channels primarily during wet-season rainfall and/or snowmelt events by sheetwash and rapid mass movements. Suspended sediment output from gidlies : seasonal and sub-seasonal patterns 3) Suspended sediment output from gullies occurs almost entirely during large rainstorm and/or snowmelt runoff events. 4) Regardless of antecedent hydrometeorological conditions, the largest runoff events transport the greatest quantities of suspended sediment. 5) By controlling replenishment of channel-margin sediment supply, antecedent hydrometeor-ological conditions influence suspended sediment output from gullies during small to intermediate runoff events. Suspended sediment output from gidlies : synoptic-scale patterns 6) Within runoff events not dominantly generated by snowmelt, Q-SSC relations exhibit clockwise hysteresis patterns due to depletion of channel-margin sediment supply during runoff events. 37 7) At very high flows, single-valued Q-SSC relations may occur in gullies with numerous hillslope sediment sources, due to accessing of unlimited channel-margin sediment supply. 8) Counter-clockwise or figure-8 hysteresis may occur during runoff events generated by snowmelt, when channel-margin sediment storage may be protected from gully-channel flows by snow cover on the rising limb but exposed to flow later in the event. 9) Within runoff events in gullies, duration of suspended sediment transport is brief relative to the duration of runoff events. Suspended sediment transport in gully-channels is almost always supply-limited. Suspended sediment transport in streams : seasonal and sub-seasonal patterns 10) Suspended sediment transport in gravel-bed streams occurs almost entirely during large rainstorm and/or snowmelt runoff events. 11) Regardless of antecedent hydrometeorological conditions, the largest runoff events transport the greatest quantities of suspended sediment. 12) The influence of antecedent hydrometeorological conditions on small and intermediate runoff-event suspended sediment transport is weaker in streams than in gullies due to variable conditions over larger drainage area and the buffering effect of in-channel sediment storage. Suspended sediment transport in streams : synoptic-scale patterns 13) During small to intermediate runoff events, Q-SSC relations exhibit weak clockwise hysteresis. During larger runoff events in which armour-layer entrainment occurs, however, clockwise hysteresis in Q-SSC relations may occur. 38 14) Suspended sediment transport during runoff events in gravel-bed streams exceeds the duration in gully channels, but is also limited in duration by supply rather than by runoff duration. Debris-torrent versus fluvial fine-sediment transport The unexpected opportunity to test a final hypothesis was provided by the occurrence of a debris torrent in one of the monitored gullies shortly after the study period. 15) Long-term fluvial sediment yield from torrent-prone gullies approaches debris-torrent yield when only the fine sediment fraction is considered. If shown to be true, these hypotheses provide the basis for qualitatively predicting patterns in fine-sediment transfer through an important sediment cascade pathway using regional hydrometeorological forecasts. Prediction of sediment transfer activity at individual hillslope sediment sources based on Hypotheses 1 and 2 is probabaly not practical using regional hydrometeorological forecasts due to site variability. However, knowledge of hydrometeorological conditions preferentially conducive to sediment transfer activity at various hillslope sediment source types should provide an adequate basis for predicting suspended sediment output patterns from gullies containing known hillslope sediment source types. Similarly, suspended sediment transport patterns in streams should be predictable based on knowledge of hillslope and gully hydrometeorological-response characteristics within the drainage basin and sediment storage characteristics within the stream channel. This study focusses on characterisation of temporal patterns of fine-sediment transfers through the hillslope-gully-stream cascade pathway using the hypotheses listed above as a basis for characterisation. For practicality reasons, emphasis was placed on suspended sediment 39 transport in the gullies and in Russell Creek. Hydrometeorological and hillslope-transfer data are rudimentary, but provide background information against which suspended sediment transport patterns can be compared. This study does not aim to provide a formal qualitative model of fine-sediment transfer activity; rather, the study presents a characterisation of temporal patterns upon which a model may be based in future work. 40 Chapter 2 Study Setting and Design This study was conducted in Russell Creek Basin, northeastern Vancouver Island, British Columbia. Russell Creek is the largest tributary of Tsitika River, which in turn flows into Johnstone Strait, approximately midway between Campbell River and Port Hardy (see Figure 2.1). The study was designed to characterise the short-term, temporal patterns of fine-sediment transfers through the sediment cascade of Russell Creek Basin under normal-regime conditions, as controlled by temporal patterns in hydrometeorological conditions. To that end, a three-level, hierarchical, nested series of sediment-transfer monitoring stations was established, to represent fine-sediment transfers in three components of the Russell Creek Basin fine-sediment cascade, namely: 1) Hillslope to gully channel 2) Gully channel to stream channel 3) Stream channel to basin outlet. Fine-sediment transfers in each of the monitored cascade components were compared to temporal patterns in hydrometeorological conditions, also recorded in Russell Creek Basin. The physical setting of each level within the sediment monitoring system is described in this chapter. 41 Figure 2.1. Location map, Russell Creek Basin. Russell Creek is located on northeastern Vancouver Island, British Columbia. 42 2.1 Nested spatial sediment-monitoring scales At the largest scale in the three-level hierarchy of nested fine-sediment monitoring sites, Water Survey of Canada (WSC) operates a flow and suspended sediment gauging station near the mouth of Russell Creek (WSC Gauge # 08HF007). Drainage area upstream of the WSC gauge is about 31 km 2 (3100 ha). At the intermediate scale, I gauged flow and suspended sediment output from two gullies within Russell Creek Basin. The gullies are referred to as Gullies "B" and "C"; their drainage areas are 12.5 and 3.0 hectares (125,000 and 30,000 m2), respectively. At the smallest scale, I monitored sediment transfers from discrete hillslope sediment sources to gully channels within the two gauged gullies; discrete sediment source areas were on the order of a few tens to hundreds of square metres. Figure 2.2 shows the locations of the WSC gauge and the two gully gauging stations within Russell Creek Basin. Figure 2.3 provides a more detailed map of the area surrounding the study gullies. The scale ratios of the largest to the intermediate study scales (Russell Creek Basin to gully catchments) are about 250:1 and 100.0:1. The scale ratios of the intermediate to the smallest study scales (gully catchments to sediment sources) are on the order of 1250:1 to 300:1. Generalising, each increasingly large study scale, from sediment source to gully catchment to Russell Creek Basin, represents an increase of approximately three orders of magnitude in spatial scale. 2.2 Hydrometeorological conditions The duration of the study period was intended to cover one annual cycle of hydrometeorological conditions; however, since flow in the study gullies is ephemeral, emphasis was placed on the wet winter season and the spring snowmelt season. The study period extended from 30 September, 1993, until 27 June, 1994. 43 Hydrometeorological conditions in Russell Creek Basin were recorded at one weather station and two snowpack measurement sites. Rainfall and air temperature were recorded at weather station "Russ2", located near Gully B at approximately 560 m elevation. Snowpack measurement sites "1" and "2" were located near the lower and upper extents, respectively, of Gully B catchment at elevations 520m and 760 m. Both snowpack measurement sites were located in clearcut areas. In coastal British Columbia, open-area snowpacks generally achieve greater wintertime accumulations, which also melt more rapidly during rainstorms and sunny weather, than do snowpacks in forests (Hudson 1995). The locations of the weather station and the snowpack measurement sites are shown in Figure 2.3. The nearest Atmospheric Environment Service (AES) weather stations to Russell Creek Basin are Port Hardy Airport (approximately 100 km to the northwest) and Campbell River Airport (approximately 90 km to the southeast). Although Campbell River Airport is the closer AES station, Russell Creek basin lies in the same climatic region as Port Hardy Airport (Klinka et al 1984). Weather data from Port Hardy Airport were used in this study for three purposes. Firstly, Port Hardy Airport data were used to estimate weather conditions in Russell Creek Basin during periods of missing Russ2 data. Secondly, the long-term (greater than 30 years) record at Port Hardy Airport allowed the study-period weather conditions to be placed into perspective. Thirdly, linking the study to a regional weather station permits examination of the study hypothesis that fine sediment transfer activity can be qualitatively forecast on the basis of regional hydrometeorological information. 2.3 Physical characteristics of Russell Creek Basin Suspended sediment yield at the WSC Russell Creek Gauging Station is the result of fine-sediment transfers within the 31 km2 Russell Creek Basin. The basin is mountainous, has a 46 temperate west-coast climate, and has been partially clearcut logged. Landscape of Russell Creek basin is shown in Figures 2.4 and 2.5. Figure 2.4. Russell Creek Basin, looking eastward from Tsitika Valley. 2.3.1 Physiography Elevation in Russell Creek Basin ranges from about 250 m at the stream's confluence with Tsitika River, to about 1700 m at the mountain summits along the basin divide. Topographically, the basin consists of a broad valley-bottom, running roughly northeast to southwest. Rugged mountains lie to the northwest and southeast of the main valley (see Figure 2.2). Russell Creek basin is underlain by two main geologic units. The valley-bottom is underlain by igneous intrusive bedrock (Island Intrusions), while the mountain peaks are composed of basaltic volcanics of the Karmutsen Formation (Geological Survey of Canada 1983). The basaltic bedrock of the mountain peaks is generally exposed, and frequent basaltic outcrops occur on hillslopes. Much of the hillslope area of the basin is mantled with colluvium, consisting of coarse basaltic rockfall deposits and debris slide deposits. Glacial till is present on most of the 47 b) Figure 2.5. View of study gullies, a) Photo taken from valley-bottom, looking roughly eastward; deeply incised Gully B in photo left, b) Photo taken from valley-bottom, looking roughly southeastward; Gully B at left, Gully C (not visible) is located on forested hillslope in photo centre-right. 48 lower hillslopes and valley bottom. The till is generally derived from the intrusive rock, consisting of granitic stones and relatively sandy matrix (Maynard 1991). 2.3.2 Cl imate The climate of northern Vancouver Island is temperate west-coast, generally mild and wet. At Port Hardy Airport, which is situated near sea level, annual precipitation averages about 1800 mm, of which only 4% occurs as snow. Winters are much wetter than summers: over 70 % of annual precipitation falls from October to March. Air temperatures are moderate, with monthly averages ranging from 2 C in January to 14 C in July and August. In Russell Creek Basin, the seasonal distribution of precipitation is likely similar to that at Port Hardy Airport. As evidence, at the WSC Tsitika River Gauging Station (#08HF004), drainage area of which includes Russell Creek Basin, all historical peak annual flows during the station's 20-year record have occurred between October and February. The three-year record at Russell Creek Gauging Station shows the same pattern. Total precipitation in Russell Creek Basin, however, may be considerably greater than at Port Hardy Airport due to orographic enhancement by the rugged relief of the basin. Also, considerable snow falls in Russell Creek basin. The transient snowpack elevation zone in coastal British Columbia typically occurs at 300 to 800 m (British Columbia Ministry of Environment, Lands and Parks and Ministry of Forests 1994), representing the valley bottoms and lower hillslopes of the basin (see Figure 2.2). Higher mountain peaks are subject to deep seasonal snowpacks. 49 2.3.3 Vegetation and land use Three elevationally controlled biogeoclimatic zones are represented in Russell Creek Basin (British Columbia Ministry of Forests 1992): the Coastal Western Hemlock zone (under 1000 m), the Mountain Hemlock zone (1000 to 1500 m), and the Alpine Tundra zone (above 1500 m). Russell Creek Basin has been partially clearcut logged by MacMillan-Bloedel, Eve River Division, beginning in the 1980s and continuing until present. Clearcuts occupy about 30% of the basin, generally on lower hillslopes and valley bottoms in the Coastal Western Hemlock zone. The area logged prior to October 1993 is shown in Figure 2.2. 2.3.4 Russell Creek channel The channel of Russell Creek is lined primarily by valley-flats; in these reaches banks are typically low. In a few reaches, the channel impinges upon steep hillslopes, where high, active slopes of glacial-till and colluvium yield sediment into the stream channel. Such reaches occur along the mainstem of Russell Creek within the first kilometre upstream of the WSC gauge, and along much of Russell Creek's main tributary, Stephanie Creek. Morphology of Russell Creek is characterised by channel bars, riffles, and pools. Bed material ranges from pebbles to boulders. Russell Creek channel morphology is partly, but not largely, controlled by LOD. 2.4 Physical characteristics of the study gullies Physical characteristics of Gullies B and C are summarised in Table 2.1 and are discussed below. Gully B has more irregular morphology and more abundant hillslope sediment sources than Gully C. Figure 2.6 provides a detailed map of Gully B. A corresponding map of Gully C did not seem to be warranted due to its morphologic simplicity. - 50 Table 2.1. Physical characteristics of Gullies B and C Gully Characteristic Study B Gully C Drainage area above gauge (ha) (1) 12.5 3.0 Area within gully crests (ha) (2) 3.3 0.4 Mainstem channel length above gauge (m) 962 356 Elevation range (m AMSL) - Gauge elevation 510 600 - Channel source elevation 820 728 Average channel gradient 0.32 0.36 Average gully dimensions (3): - crest-to-crest width, "w" (m) w - 11 (0 - 440 m u/s) w~ 12 w~ 60 (440 - 840 m u/s) w~ 15 (840 - 962 m u/s) - crest-to-channel depth, "d", (m) d ~ 2 (0 - 440 m u/s) d~2.5 d ~ 26 (440 - 840 m u/s) d ~ 5 (840 - 962 m u/s) Clearcut area (% drainage area) 99 0 Sediment source types: - channel banks Y - LOD wedges Y Y - active sidewall slopes Y - failure scars Y - road cutbanks Y - road fillslopes Y - road surfaces Y -Sediment origins: - colluvium Y Y - glacial till Y - weathering bedrock Y -Silt & clay abundance in sediment sources 15-25 2-10 (as % of sub - 2mm material) (4) cn-301 .wb1 Notes: 1. Estimated from 1:5000 topographic maps. 2. Estimated from ground surveys. 3. In Gully B, dimensions are averages for gully sections, defined by distances upstream of gauge. See Appendix C-1 for gully survey data. 4. See Appendix C-2 for particle-size distributions of bulk-sampled sediment sources. 51 2 on c o Ml T3 ! a. o cj 1) D . C L o o IS H3 o o e CJ E •3 CJ to •a P J l UJ U y u l = v, 6 "3 2 — in 5 CJ ca "> § K E TO to •Ms LO TO LO J O S 3 u u cs n * wi o t3 C CD OJ CJ ca T3 LO CJ C C ~ TO o 3 .5 3 73 73 x f !y fe "fi 5 t S E . . v> is ) o cj </f ^ S 3 XS . CJ T3 O u <£ 3 LO <" 1-2 3 Z u CJ o eo c> o" CN CO XI o TO <L) O rs I CJ cj ,TO t l 3 LO T 3 L* TO J O 3 u TO J O 3 o J 3 Xi 2 2 J ° ° g U Q b o o « cs eN cn i—, >n CO CO CO H H H 2 2 2 < co co r n </-> CO CO CO 5 .a 13 o 3 TO I 8 I < 8 £ '-5 -5 —' f i -3- v i r- oo h n, h b n. h LO TO o » LO J O CJ T 3 < £ CN o o' o. o o VI CN 2 a. o c o _l m >» "5 O s 10 (ui) uo|)eA9|3 53 o 60 e o 3> - a c 00 CL) o l_ 3 O oo +-* G CD oo "<3 cd JS OJ a. o 2 PQ >> vo cs eu a 2.4.1 Gully situation As shown in Figure 2.2, Gullies B and C are situated approximately 1 km apart, on lower hillslopes in Russell Creek Basin. Aspects of Gullies B and C are approximately west-northwestward and north-northwestward, respectively. The elevation range covered by Gully B is 510 to 820 m, and by Gully C is 600 to 728 m. Therefore, both gullies are situated within the coastal British Columbia transient snowpack elevation zone and the Coastal Western Hemlock biogeoclimatic zone. Gully B has been entirely clearcut logged, and is now vegetated primarily by regenerating Western Hemlock (Tsi/ga heterophylla) under 15 years in age. Gully C is entirely forested by old growth Western Hemlock and Sitka Spruce {Picea sitchensis). 2.4.2 Gully morphology Gully morphology was determined from 1:5000 scale maps produced by MacMillan-Bloedel and from ground surveys. Drainage area estimates made from the maps are approximate since contributing areas from unchanneled slopes above the gully-channel sources and adjacent to gully crests were difficult to identify. Gully widths and depths refer to dimensions of surveyed gully incisions. ' Gully B width and depth are not only greater than in Gully C, but they are also more irregular. Gully C is incised into a regular, smoothly sloping hillslope. In Gully C, width and depth average about 12 m and 3 m, respectively, along the length of the gully. Gully B, on the other hand, begins on a moderate-gradient hillslope, then becomes deeply incised in a hillslope bench. The bench ends abruptly at a sharp, convex slope break, along which several debris slide scars converge into the gully (see Figure 2.6). Old dormant debris-scar depressions indicate that Gully B sidewall failures occurred long before logging. However, timber removal from the unstable depressions may have enhanced debris sliding incidence. The hillslope is steep below the 53 convex slope break, then gradually moderates toward the valley-bottom. In the upper and lower sections of Gully B, width and depth average about 10 to 15 m and 2 to 5 m, respectively. In the 400-metre long, deeply incised section, however, width and depth average about 60 m and 26 m, respectively. Gully B channel length is the longer of the two: 962 m, as opposed to 356 m. Despite the more irregular hillslope gradient at Gully B, the average gradients of Gullies B and C are similar — 0.32 and 0.36, respectively. 2.4.3 Sediment sources The main sources of hillslope-to-channel sediment transfers in Gully C are channel banks. The banks are mainly composed of sandy-matrix colluvial deposits, and are typically undercut by about 30 to 40 cm along the length of the gully channel. The lower banks are eroded by high gully-channel flows, but the upper banks are held together by the dense root mat of the old growth forest. No other exposed soil surfaces were observed within the gully (see Figure 2.7). Gully B, on the other hand, has numerous exposed soil surfaces which contribute sediment to the channel (see Figures 2;8 and 2.9). Eight sidewall failures and five actively eroding sidewall slopes exist in the incised middle reach of the gully. In addition, the channel is crossed by logging roads at four points. Logging roads may produce sediment by means of road surface wash, cut-bank erosion, ditch erosion and erosion of sidecast material. The roads crossing Gully B channel were partially deactivated prior to the study period; road-related fine-sediment transfers to the Gully B drainage network appeared to be minor. The channel of Gully B is lined by channel-margin sediment storage along much of its length in the deeply incised section; below this section, the channel is typically wide and shallow due to coarse sediment deposition from the sediment sources above. Therefore, little undercutting of banks occurs in Gully B. The sediment 54 Figure 2.7. Gully C morphology. Bed material is dominated by stable, mossy, lag boulders; banks are vegetated and undercut; sidewalls are vegetated and stable. transferred into Gully B contains a higher proportion of silt and clay than does the sandy bank material of Gully C (see Table 2.1). 2.4.4 Channel morphology The channel of Gully C is characterised by lag-boulder controlled, step-pool morphology, with additional step-support provided by small LOD structures. The steps and pools are relatively small and frequent (one cycle per 1 to 6 m). Channel banks are typically undercut by 30 to 40 cm. 55 Figure 2.8. Gully B sidewall failures, a) Looking southward at sidewall failures on left gully wall, Road TS120D in background. Failures seen, from left to right, are F8, F7, and F4. b) Failure F5 on right gully wall. 56 Figure 2.9. Gully B active sidewall slope and road cutbank. a) Looking down gully channel, active sidewall slope S3 at left, b) Road cutbank at TS120 crossing. Log at photo left-centre is part of log culvert. Sediment source monitoring site 6. 5 7 In the upper and lower sections of Gully B, steps are somewhat higher and less frequent than in Gully C, and are dominated by logging slash. Coarse sediment wedges occur behind most slash jams. In the deeply-incised middle section of Gully B, larger, less frequent LOD jams occur. Periodic sidewall debris slides have initiated debris torrents which have scoured much of the channel to bedrock and created jams of coarse slash with associated sediment wedges. The largest L O D jam and sediment wedge occur about 50 m upstream of TS120C road crossing (see Figure 2.10). This jam is 19 m in height, and the sediment wedge is 38 m in length by 17 m in width. The L O D appears to be draped over a bedrock face, so the sediment wedge thickness is considerably less than the 19-metre height of the jam. Assuming an average sediment depth of 2 metres and a triangular planimetric shape, the volume of sediment stored within the wedge is approximately 650 m3. Between LOD jams in the middle section of Gully B, the scoured channel is often lined by channel-margin sediment storage compartments created by the various hillslope sediment sources. 2.5 Study sediment sources Within Gully B, five discrete sediment sources were selected for monitoring hillslope to gully-channel sediment transfers. The five sources consisted of two failure scars, one actively eroding sidewall slope, one road cross-ditch bank, and one road cutbank. In addition, the large sediment wedge above Road TS120C (Figure 2.10) was monitored as an in-channel source of fine sediment. Locations of the six study sediment sources are shown in Figure 2.6. The failure scar surfaces are composed of relatively impermeable glacial till. Shallow veneers of sandy-matrix material derived from the upslope soil layers periodically accumulate and are flushed off the till surfaces. The actively eroding sidewall slope and road-related banks are 58 Figure 2.10. The big LOD jam and associated sediment wedge in Gully B, located above road TS120C. a) Looking at the LOD jam from below (TS120C); jam height is 19 m. b) Looking at the sediment wedge from upstream; the wedge is 38 m long by 17 m wide. 5 9 composed of sandy-matrix material. Texture of the large sediment wedge ranges from cobbles and boulders at the apex to fine sand amongst the L O D pieces in the jam. 2.6 Summary In summary, this study was designed as follows. Hydrometeorological conditions (air temperature, rainfall, and snowpack) and their effect on normal-regime, fine-sediment transfers through the sediment cascade were monitored in Russell Creek Basin, from September 1993 until June 1994. Specifically, the activity of hillslope to gully-channel fine-sediment transfers was monitored in one gully (Gully B). The suspended sediment output from Gully B, and from another gully (Gully C) lacking hillslope sediment sources other than undercut banks, were monitored. Finally, suspended sediment output from the entire Russell Creek basin was monitored. The three nested spatial scales of fine-sediment transfer monitoring allows a characterisation to be made of the temporal patterns of fine-sediment transfer along an entire basin cascade pathway, from hillslopes to basin outlet. Temporal patterns of hydrometeorological conditions were used as the reference to which sediment transfer patterns were compared. The inclusion of two very different gullies from the same area was meant to provide insight into the degree of spatial variability which occurs in temporal patterns of fine-sediment transfers in gullies. The results of this study should not be interpreted as a direct comparison of a logged gully versus an unlogged gully. At the minimum, such a comparison would have required a pre-logging calibration of sediment transfer characteristics between the two gullies. Such a calibration was not conducted. 60 Chapter 3 Methods This chapter discusses the methods by which hydrometeorological and fine-sediment transfer data were collected in this study. Collection methods for hydrometeorological data and hillslope to gully-channel sediment transfer data were rudimentary, designed to provide general information on conditions controlling suspended sediment transport in the study gullies and Russell Creek. In contrast, the suspended sediment transport data were collected at high resolution to provide a detailed view of temporal patterns. 3.1 Hydrometeorological conditions 3.1.1 Weather Rainfall and air temperature data were collected in Russell Creek Basin at weather station "Russ2" by Scott Davidson, as part of concurrent research on hillslope groundwater response to heavy rainfall. Rainfall was measured by a tipping-bucket rain gauge and air temperature was measured with a thermistor, both recording on a Unidata data logger at 15-minute intervals. Both instruments were lab tested before field installation. The rain gauge was subject to icing and snow cover, resulting in some data loss, but blockages were cleared during field visits, which occurred at three-week intervals on average. Snowfall was not measured. 61 3.1.2 Snowpack depth Snowpack depths were periodically measured at two sites, providing information on net snow accumulation and ablation at approximately three-week intervals. Snow depth was measured manually with a metre-stick. The snowpack depth data are only indicative. Three factors limited data accuracy. First, depth rather than water-equivalence was measured for simplicity. Therefore, snowpack density changes cannot be distinguished from mass changes. Second, snowpack depths are based on one or two measurements only, rather than the many points required to represent the diversity of depths that can occur due to microtopography and microclimate within a small area. Third, the measurement interval was irregular and occasionally exceeded one month. Snowpack depth data were recorded only to identify gross patterns of snowpack accumulation and ablation. Interpretation of the snowpack data is further complicated by the location of both measurement sites in clearcuts, near or within the catchment of Gully B. The snowpack variation observed over time at the two sites probably represents the conditions of the snowpack in Gully B catchment moderately well. In forested Gully C, however, maximum snowpack accumulations can be expected to have been less, and snowpack ablation during rainstorms and sunny weather to have been less rapid. Field observations confirm that mid-winter snowpack depths in the forested catchment were indeed less than at equivalent elevation in the clearcut. 3.1.3 Synoptic-scale weather system classification As part of the plan to link Russell Creek Basin fine-sediment transfers to regional weather, synoptic-scale weather systems affecting northern Vancouver Island during the study period were classified by analysis of regional. weather maps (Environment Canada 1995). Weather systems were classified as follows: 62 1) cyclonic storm system: consisting of one or several rapidly consecutive fronts associated with a low-pressure cell; 2) inter-cyclonic lull; 3) cP or cA (continental polar or continental arctic) anticyclone: high-pressure ridge associated with anticyclonic cell originating in northern Canada; 4) subtropical anticyclone: high-pressue ridge associated with anticyclonic cell originating in southwestern U.S. or adjacent Pacific Ocean. 3.2 Hillslope to gully-channel sediment transfers Two techniques were employed to monitor temporal patterns of sediment source activity at the hillslope to gully-channel transfer sites, namely erosion / deposition pins and sediment trapping. No attempt was made to equate rates of sediment transfer between the various methods or sites; only the temporal pattern of sediment transfers, at a given site, as measured by a given method, can be inferred. 3.2.1 Erosion / deposition pins Sites descriptions Erosion / deposition pins were used to monitor sediment transfer activity at Sites 1 (sidewall failure scar), 2 (active sidewall slope), 4 (LOD-jam sediment wedge), 5 (cross-ditch bank) and 6 (road cutbank). The pins at Site 4 consisted of five rebar rods, each about 1 metre long and 10 mm in diameter, inserted vertically into the surface of the in-channel sediment wedge. Repeated measurement of exposed pin length provided a record of fluvial erosion and deposition on the wedge. The pins at the other four sites consisted of spikes, 25 cm long and about 8 mm in 63 diameter, inserted perpendicularly to the surface of the sediment source site. Once again, exposed pin length was repeatedly measured, in this case providing a record of erosion and deposition by the various hillslope processes operating at the sites. Eight pins were installed at each of Sites 1 and 2, five pins at Site 5, and six pins at Site 6. At all sites, the pins were not inserted all the way into the ground; rather, significant portions were left exposed initially to allow measurement of deposition if required. Data interpretation Erosion / deposition pin measurements at Sites 1, 2, 5, and 6 represent transfers of sediment from upper banks and slopes to channel-margin storage compartments. Figure 3.1(a) shows the erosion / deposition pins at Site 6 (road cutbank). The pins are located in the near-vertical bank face; erosion rates refer to sediment transfers from that face to the loose-sediment apron along the base of the bank, not directly to the channel. However, the size of the apron, and the addition of fresh fine sediment to the surface of the apron, determines the nature of the sediment supply available to the channel flow. As the apron becomes larger, it extends further into the channel, given a constant angle of repose. The addition of fresh fine sediment to the surface of the lower apron is required to replenish the fine-sediment supply, as the lower apron surface material may become coarsened by winnowing of fines by gully-channel flow. The erosion / deposition pin measurements at Site 4 (sediment wedge) represent fluvial erosion and deposition within the gully-channel. In other words, these measurements represent transfers along the gully channel, as opposed to transfers to channel-margin supply sites for potential future transport down-gully. Site 4 is shown in Figure 3.1(b). 64 Figure 3.1. Erosion / deposition pins, a) Site 6 (road cutbank). The pins are located in a near-vertical eroding face; a depositional storage apron covers the lower bank, b) Site 4 (LOD-jam sediment wedge). Material texture changes from coarser (boulder / cobble) at the wedge apex to finer (sand / silt) at the wedge snout. Two pins are visible; three more pins are located in the fine material amongst the LOD (similar to foreground) 65 Error sources The erosion / deposition pins provided an easily obtained and analysed record of sediment source activity; however, several problems are associated with their use (Haigh 1977, in: Toy 1983). These problems include soil structure disturbance; operator site disturbance; impedence by pins of surface water / sediment movement; different behaviour of pins compared to soil, such as preferential frost heave of steel pins; bending, removal, or burial of pins; difficulty in differentiating soil surface disturbances (eg: expansion due to needle ice growth, before erosion of loosened material) from the sediment removal / addition of interest (erosion / deposition); and measurement error. Attempts were made to minimise the effect of these error sources. Soil structure disturbance was inevitable, but the pins were pounded into the soil carefully to minimise disturbance. Operator disturbance was minimised by not walking beside or upslope of pins; pins were installed in horizontal rows across sediment source sites to allow this. The pins appeared to modify surface sediment and water movement; sediment accumulation was commonly observed on the upslope side of pins, and scour by channelised overland flow rivulets was commonly observed on the downslope side of pins. Pin exposure was measured on the side of the pins in an attempt to average the effects of these two phenomena; the right-hand-side of the pins was chosen as the convention. Occasional frost heave of pins out of the soil occurred; the heaved pins were gently pushed back into their holes before exposure length was measured. Pins occasionally were buried, fell out of their holes, or were carried downslope with a minor mass movement event. Burial depth was measured when possible, and this was simply represented as a negative pin exposure'length. Pins that fell out or were lost were replaced in approximately the same location, and a new initial exposure length was measured. When pins 66 were incorporated in minor mass movements, the depth of failed material was sometimes possible to determine by the depth of small scarps bordering the failure. Surface disturbance by needle ice growth created a very difficult data interpretation problem. Although needle ice could be detected visually when present during pin-exposure measurement, its length was impossible to measure without disturbing the delicate needles and the sediment layer perched on the needle tips. Figure 3.2 shows an example of ice needles that developed in road cutbank sediment in Russell Creek basin. Pin-exposure lengths recorded during needle ice presence appear as deposition in the data record. Measurements in the data record that were subject to needle ice influence were indicated. Measurements made prior to, and following, pin measurements subject to needle ice were compared for net surface position change. Figure 3.2. An example of needle ice growth in road cutbank sediment in Russell Creek Basin. 67 3.2.2 Sediment trapping Sediment trapping was utilised only at Site 3 (failure scar). The trap consisted of a geo-cloth mesh weir, blocking the channelised transport zone at the base of the failure. The mesh was designed to allow water and clay-sized sediment to pass through, but to trap silt-sized and coarser material. The weir was about 60 cm high and 90 cm wide. The trap was installed on 18 November, 1993, and had completely filled with unsorted sediment (fines to boulders) by 6 December, 1993. The sediment trap was not maintained afterwards. 3.3 Suspended sediment transport in gullies Gauging stations were operated near the mouths of the two study gullies to record output of suspended sediment. Suspended sediment transport was calculated as the product of water discharge times suspended sediment concentration (SSC). Instrumentation at the gauging stations recorded water stage, from which discharge was obtained, and water turbidity, from which suspended sediment concentration was obtained. 3.3.1 Gully gauging stations V-notch weirs were constructed at each of the gauging stations to provide channel control and increase the range of water depths experienced, thereby increasing measurement resolution. Instruments were installed in the pools upstream of the weirs. Instrument data were recorded on CR21 data loggers, which were housed in Coleman camping coolers near the weirs. Data collection interval was 15 minutes, which provided a compromise between obtaining sufficiently high resolution data to allow synoptic-scale analysis, while not overloading data-storage capacity between field visits which sometimes occurred less frequently than once per month. Photos of the 68 gauging stations are provided in Figures 3.3 and 3.4. Site sketches of the gauging stations are provided in Appendix C-3. 3.3.2 Stage and discharge Stage measurement method Stage was recorded with 1-metre Unidata capacitance sticks. Each weir also had a staff gauge mounted on the pool side (upstream side) of the weir faceplate. Recorded capacitance stick readings were related to staff gauge stages during field site visits, so that each gauging site had a physical water level reference. The physical reference was required as a check on the electronic water level recording system. Staff-gauge stages were then converted to V-notch heads, which were the reference water level values used for water discharge calculation. Details of conversions from capacitance stick readings to staff gauge stages to V-notch head values are provided in Appendix B - l . Stage measurement errors The accuracy, resolution, and linearity of the Unidata capacitance sticks were stated in the manual to be 5 mm, 4 mm, and 1 mm, respectively. In laboratory calibration tests, the capacitance sticks used in this study did not meet these specifications. The capacitance sticks were lab calibrated by adjusting offset and gain at 0.100 m and 0.500 m water depth. Then capacitance stick readings produced by the two units used in the study were simultaneously compared to actual water depths while incrementally filling a bucket to depths between 0.100 and 0.500 m. The capacitance stick used at Gauging Station B responded linearly to changing water depth, while the capacitance stick used in Gauging Station C did not. 69 The non-linearity of the Station C capacitance stick seemed to be systematic, and amounted to a maximum of 8 mm between the controlled calibration points. The data recorded at Gauging Station C were adjusted to compensate for the non-linearity. The capacitance stick readings were found to be positively related to voltage supply, below a critical threshold of about 4.5 volts. The stated voltage supply requirement of the capacitance sticks was 5 volts. However, 6-volt lantern batteries were used to power the capacitance sticks; these batteries were usually replaced before voltage was drawn down to 4.5 volts. The capacitance stick readings also showed a slight positive relation to both air and water temperature. During independent air and water temperature changes of 9 C (ie. either air or water temperature was varied while the other was held constant), capacitance stick readings varied by 3 to 5 mm. Instead of determining temperature correction functions, these errors were simply incorporated into the uncertainty of the measurements. Some error reduction was gained, however, by re-establishing capacitance-stick versus staff-gauge stage relations every time the batteries were changed (four to five times during the nine-month study period). Air and water temperature ranges between battery changes were likely much less than the ranges during the entire study period. Discharge measurement methods Water discharge was manually measured, at known stages, by current metering and by bucket retention. Bucket retention was used whenever flows were low enough to permit it (less than about 10 l/s). A plastic garbage pail with a capacity of about 20 litres was held under the V-notch for a length of time recorded by stopwatch. The arbitrary water volume collected in the bucket was 72 determined by measuring the water level in the bucket relative to the bucket rim, and then comparing the measured rim-to-water distance to a calibration table of water level versus water volume. A Marsh McBirney electromagnetic current meter was used to measure discharge when flows were too great to be collected by the bucket retention method; this was required only at Gauging Station B. Regular, straight channel segments were created for current meter discharge measurement by rearranging bed material, always within a few metres downstream of the weir. The measurement site always disappeared between measurement times due to deposition of bed material excavated from a growing scour hole beneath the weir notch. The current meter was used to measure water velocity at three to five points in a channel cross-section; velocity was then multiplied by water depth and cross-section segment width to obtain discharge. Discharge measurement errors Errors in the measurement of discharge by the bucket retention method could have arisen due to weir leakage, and notch-to-bucket spillage. In either case the true water discharge would have been underestimated. No attempt to quantify these errors was made. However, qualitative observations of water ponding in the weir pool even under very low flow conditions indicate that weir leakage was minimal. Notch-to-bucket spillage was also observed to be negligible. Errors in the measurement of discharge by the current meter method could have arisen from instrument inaccuracy and imprecision, intragravel flow, and channel irregularities. Stage-discharge correlation methods Water discharge was measured at a range of known stages, by the methods discussed above. A rating curve was drawn for each gauging station to relate stage and discharge. In •73 addition, standard formulae relate water head in a V-notch weir to flow through the notch. Therefore, the rating curve and the 90-degree V-notch formula provide checks on one another. In general, measured discharge agreed well with formula-predicted discharge at low flows (discharge less than 3 l/s at Gauging Station C, and less than 5 l/s at Gauging Station B); however, the formula underestimated measured discharge at flows above these flow thresholds. Stage-discharge correlation errors Disagreement between measured and predicted discharge may be due to either discharge measurement errors (discussed previously) or deviation of weir design from specifications required to meet assumptions of the predictive formula. Weir leakage results in disagreement between measured and predicted flow, if measurement is made by current meter, because the sub-weir flow is measured in the gauging section below the weir; in other words, the formula underestimates discharge. If the bucket retention method is used, however, the sub-weir flow is not detected at all, and no disagreement results (ie. the importance of this error must be determined by some other means, as it will not affect the rating curve check). Disagreement also occurs if the stilling pond behind the weir is hot large enough; the predictive formula assumes a calm, still water surface behind the V-notch, but "shooting flow" occurs if the pond is too small. In this case, the formula underestimates discharge, with the degree of underestimation increasing rapidly with increasing flow. The formula assumes that the water flowing through the V-notch falls freely once through the notch, otherwise "drowned flow" occurs. The V-notch crest is assumed to be a sharp edge; flatness of the crest introduces drag to the water flowing over it, slowing the water, and resulting in overestimation of discharge. The underestimation of discharge above the respective flow thresholds at Gauging Stations B and C was probably due to shooting flow through small weir ponds. To account for 74 this error, the standard weir rating formula was used below the shooting flow thresholds, and a new rating formula was fitted to measured stage-discharge correlated points above the thresholds. The correlated stage-discharge data, and the rating formulae based upon them, can be found in Appendix B-2. The rating curves are shown in Figure 3.5. 3.3.3 Turbidity and suspended sediment concentration Turbidity measurement methods Water turbidity was recorded with D & A OBS (Optical Back-Scatter) sensors. The OBS sensors emit a constant beam of infra-red radiation and measure the amount of the radiation reflected back to the instrument. Thus, the instruments measure the amount of scattering by suspended particles in the water. Good correlation between turbidity and suspended sediment concentration has been observed in many studies (for example, Truhlar 1978, and Beschta 1980). Turbidity measurement errors Besides suspended solids concentration, water turbidity can also be affected by the following water properties: colour, dissolved material content, sunlight, entrained air bubble presence, and the presence of nearby radiation obstacles, such as boulders. In most mountain streams, however, the suspended solid concentration is the dominant turbidity-controlling property; more specifically, suspended sediment (ie. inorganic suspended solids) is the main contributor to water turbidity (Beschta 1980). Even if suspended sediment concentration is the main factor controlling turbidity, the particle size distribution of the suspended sediment can affect the correlation. Turbidity is inversely proportional to median suspended particle size at a given suspended sediment 75 Gauging Station B Notch Head, H (cm) • Measured Q — Q = 4.5 H 3 0 Q = 1.4 H " Gauging Station C Notch Head, H (cm) • Measured Q — Q = 5.0 H 2 J > S Q = 1 . 4 H 2 5 -Fig. 3.5. Head-discharge rating curves for Gauging Stations B and C. a) Station B. b) Station C. 76 concentration (Holstrom and Hawkins 1980). As suspended particle size increases, at a given concentration, the constant mass of the suspended material is concentrated in fewer, larger particles, which interfere less with radiation passing through the water. Therefore, the turbidity-SSC relation can be strongly correlated, yet temporally variable, at a gauging site, due to temporal variations in suspended particle size distribution. Such temporal size distribution variations can result from flow fluctuation, and consequent competence fluctuation, as well as from temporal variability of sediment supply conditions. Turbidity-SSC relations can also vary spatially, due to variability of sediment supply characteristics and sediment transport competence. Turbidity readings may also be affected by algal growth on the instrument face. The OBS sensor faces at Gauging Stations B and C were cleaned twice during the study period, in January and May 1994. After both cleanings, turbidity readings were the same as before cleaning; therefore, it can be assumed that algal growth on OBS sensor faces had negligible affect on turbidity readings in this study. SSC measurement methods SSC was measured at Gauging Stations B and C by obtaining stream-water samples, filtering them through 1.2 micron mesh filters, and weighing the dried filtered material. Organic composition was not determined, so SSC in this study refers to suspended solids concentration (ie. inorganic and organic sediment). The stream samples were obtained by two methods. At both gauging stations manual grab samples were collected in 500-ml plastic bottles, during field site visits. An ISCO 3700 automatic pumping sampler was also used to collect 800-ml samples at programmed time intervals. The ISCO sampler was used only at Gauging Station B because of difficult access to Gauging Station C. 77 SSC measurement errors • , ' • Potential SSC measurement errors include sample contamination by use o f dirty bottles, insufficient suction velocity in the I S C O sampler intake hose, and contamination o f sample filters in lab analyses. The ISCO suction velocity was calculated and found to exceed stream velocity. Other error sources were minimised by careful attention to cleanliness. 7*urbidity-SSC correlation methods Turbidity-SSC correlation was performed by comparing: a) SSC in stream-water samples versus simultaneously recorded turbidity at the gauging stations; b) laboratory-controlled SSC versus turbidity. The use o f field data from the gauging stations is the preferable method. However, insufficient stream samples were obtained during major sediment transport events to determine adequate SSC-turbidity relations for the two gauging stations. The laboratory calibration method was used to supplement the field-data relations. Sediment was collected from three sediment sources in Gully B and two sources in Gully C, sieved to remove material coarser than 125 microns, and added incrementally to a large bucket o f water in the laboratory while measuring turbidity. The lab calibrations determined did not differ noticably between gullies, so a single average turbidity-SSC relation was used for both gully gauging stations. The lab calibration agreed moderately well with the field calibration, and the goodness of agreement did not appear to differ between gauging stations. See Appendix B-3 for turbidity-SSC correlation data and computation of resultant relations. The turbidity-SSC relations are shown in Figure 3.6. 78 a) Turbidity -.SSC Relations, Lab Calibration, Gullies B and C 400 300 + s I-z £ 2 0 0 T> '.a 5 I-100 Max. turbidity recorded at gauging stations \— — U J i t -— v - * — : : — • j—Range displayed in (b) j^^F^^i^^ ' I i . . . ! 1 \ L^^^^xr.^.....\ i..: i i j !.... : 1 f " 1 t -i i ^ j i— i i i A— - i - j - i 50 100 150 SSC (mg/1) 200 250 — Gully B - - Gully C Average 300 b) Comparison of Lab Calibration Relation to SSC Samples, Station B and C 8 10 12 SSC (mg/l) 14 16 18 + Stn. B samples n Stn. C samples - — Lab calibration 20 Fig. 3.6. Relations between turbidity and SSC at Gauging Stations B and C. a) Lab calibration relations using sediment from three sediment sources in Gully B and two sources in Gully C. Average of the five relations was chosen for use at both Gauging Stations B and C. b) Field SSC samples compared to average lab calibration relation. 79 Turbidity-SSC correlation errors Errors in SSC-turbidity correlation using field data arise from the temporal variation of suspended sediment particle size distribution. Ideally, turbidity-SSC relations would be determined for every sediment transport event; however, not enough stream samples were collected to permit this. At a much shorter time scale, high-frequency fluctuations in SSC during runoff events may result in poor turbidity-SSC relations if turbidity recording duration differs significantly from SSC sampling duration. Errors in turbidity-SSC relation using lab data arise from differences between the sediment used in the lab and the actual suspended sediment in the gully-channels. Given the spatial and temporal variation in suspended sediment characteristics discussed previously, the use of a single calibration relation for all runoff events in two gullies will lead to significant error. However, major temporal patterns of suspended sediment transport, in which SSC ranges over one or more orders of magnitude, can still be confidently identified. 3.4 Suspended sediment transport in Russell Creek Data collection methods performed by WSC at their Russell Creek Gauging Station were similar to those used at the gully gauging stations. Stage was recorded by means of a pressure transducer and chart recorder. Discharge was measured by current meter gauging. Fifteen-minute average OBS turbidity readings were recorded on Unidata data loggers. Suspended sediment samples were obtained by manual depth-integrated sampling, as well as by ISCO automatic pumping sampler. I obtained raw turbidity and SSC sample data from WSC. The field-based turbidity-SSC relation was very poor (see Figure 3.7), probably due to turbidity and SSC sampling duration differences. Therefore, the lab calibration relation developed for the gully gauging stations was used at Russell Creek Gauging Station as well. Particle-size distributions of six SSC samples 80 collected during a high-turbidity runoff event at Russell Creek Gauging Station show that 73 to 91 % of the suspended material was finer than 125 microns, the truncation size used in the lab calibration. Results presented in Chapter 6 (Figure 6.8) show that the use of the lab-calibration formula at Russell Creek Gauging Station is not unreasonable. 81 a) Russell Creek SSC and Turbidity Samples Identified by Runoff Event 500 400 — 300 T J D 200 100 - © ..Q....Q 1 c . .o. 5 o ,. Q-C 3-.-... o o ...P o -t-i> ° a _ : +..... + o -1-<3 H 50 100 150 200 SSC (mg/1) 250 300 • Event 1 + Event 6 " Event 8 o Event 11 350 b) Comparison of Lab Calibration Relation to Russell Creek SSC Samples 100 150 200 SSC (mg/1) 250 300 350 SSC samples — Lab calibration Fig. 3.7. Relation between turbidity and SSC at Russell Creek Gauging Station, a) SSC samples identified by runoff event, b) Russell Creek field data compared to turbidity-SSC relation from gully gauging station lab calibration. 82 Chapter 4 Hydrometeorological Conditions and Hillslope to Gully-Channel Sediment Transfers Results of hydrometeorological and hillslope to gully-channel sediment-transfer data collection are presented in this chapter. 4.1 Hydrometeorological conditions Hydrometeorological conditions during the study period are presented in Table 4.1 and displayed graphically in Figure 4.1. The study period was subdivided subjectively into wet and dry seasons, and the wet season was further subdivided into sub-seasons defined by the alternating occurrence of cP anticyclonic conditions and maritime cyclonic conditions. Hydrometeorological conditions during the seasonal and sub-seasonal periods are summarised in Table 4.2. Runoff events refer to independent high discharge events at the gully gauging stations. 4.1.1 Observations Figure 4.1 illustrates the relatively well-defined onset of the wet season in mid-October and the absence of large rainstorms after mid-April. During most of the wet season ~ from mid-November until late March ~ daily fluctuation of air temperatures across the freezing point was common. Two cP anticyclones affected northern Vancouver Island during the study period. Russell Creek Basin air temperatures during the first cP episode in late November were the coldest of the study period; the second cP episode was considerably milder. Snowpack 83 2 o I-* > m Ul c 3-o 3 a "8 a. | ? 3 ft ° 0). § cr a' CQ (fl TV (B X o_ ni "cl T. cr 3 3 — 3 3 <= $ 9 <fl 2 8 8' ? 8 3 3 Q. S I x 3 a. s. j? i V -o CU »* sr x - 01 g a S" > (fl ^ R > °- m S 3 CB CL O CT (ft 3 O. Si ST 8 21 - 26 Nov. 27 Nov. - 01 Dec. 02 - 03 Dec. 04 - 06 Dec. 07 - 08 Dec. 09 Dec. 10 Dec. 11 -14 Dec. 15-27 Dec. 28 - 29 Dec. 30 Dec. 31 Dec. - 04 Jan. 05 Jan. 06-13 Jan. 14 Jan. 15-17 Jan. 18-20 Jan. 21 - 23 Jan. 18 Oct. 19 Oct. 20 - 23 Oct. 24 - 26 Oct. 27 - 28 Oct. 29 Oct. 31 Oct. 01 - 02 Nov. 03 - 04 Nov. 05 Nov. 06-10 Nov. 11 Nov. 12-13 Nov. 14-15 Nov. 16-18 Nov. 19-20 Nov. 10-23 Sep. 24 Sep. 25 Sep. - 04 Oct. 05 - 06 Oct. 07-15 Oct. 16 Oct. 17 Oct. j Period (Dates, 93/94) Wet season, Sub-season II Wet season, Sub-season I Dry season Temporal sc; Seasonal and , sub-seasonal I i | to i c o o o o c o o o O o c o o o G 1 1 1 x l I H r W I W r ( n w i w i o | O o O c o o c o o c o o c o o c o o c o o c o o C O r - C O X C O X C O X C O X C O X C O X C O X C O CO O CO O CO i 1 X CO X CO x 1 1 ale classificat ons Synoptic-scale weather system (1) +6/-9 +5/-1 +8/0 +5/-4 +4/-1 +3/-1 +3/0 +4/-2 +7/-2 +6/+1 +5/+1 +6/0 +4/0 +9/-1 +12/+5 +7/+3 +10/+2 +9/+1 +14/+2 +10/-2 +18/+5 +16/+3 +15/+3 +18/+6 +14/+6 +14/+1 +19/+1 +14/-1 +9/+4 +10/-1 +9 / -1 +10/-1 +6/0 +7/0 +7/-1 +22/0 +19/8 +29/+2 +19/+3 +16/-1 +13/+3 +14/+2 (a) Air temp, range (C) m £ i D 0 0 ^ - k 0 g M 0 1 ° o ' 0 £ ° ( o N) O O CD O CO Hydrometeorologii (b) Rainfall (mm) CO CO CO CO CO CO CO to :al conditions (2) (c) Snowfall occurrence 10/-18/63 0/23 0/0 0/10 (d) Snowpack depth (cm) S <D oo ~j o> cn .b. CO M Runoff Event ft 19-24 Apr. 25 - 29 Apr. 30 Apr. - 04 May 05 -11 May 12-15 May 16-24 May May 26 - 27 May 28 - 29 May 30 May 31 May 01 - 09 Jun. 10-13Jun. 14-27 Jun. 24 Jan. - 07 Feb. 08 Feb. 09-10 Feb. 11- 12 Feb. 13 Feb. 14- 15 Feb. 26 Feb. 27 Feb. 28 Feb. - 04 Mar. 05 - 08 Mar. 09-10 Mar. 11 Mar. 12- 13 Mar. 14 Mar. 15- 20 Mar. 21 - 26 Mar. 27 - 29 Mar. 30 Mar. - 01 Apr. 02 Apr. 03-10 Apr. 11 Apr. 12-14 Apr. 15-18 Apr. Period (Dates, 93/94) Dry season Wet season, Sub-season III Tempora sc; Seasonal and sub-seasonal M i l l I i I I I I £ SP8I, ale classifications Synoptic-scale weather system (1) +17/+1 +24/0 +18/0 +30 / +4 + 18/0 +29 / +2 +17/+5 +16/+2 +18/+4 +19/+3 +21 / +5 +24 / +3 +21 / +5 +24/+4 +10/-5 -2/-5 +5/-2 +7/-1 -1 / -2 +7 / -3 +9/-7 +3/+1 +4/+1 +13/-2 +14/-4 +12/-1 +9/-3 +13/-1 +14/-2 +8 / -1 +23 / -5 +25 / +3 +20 / +2 +9/+5 + 18/-1 +7/+1 +8/0 +23 / +5 -(a) Air terr p. range (C ^ M ^ O ^ N M O ^ - O W U C I w o - c D U ° ° " c o 0 r o ^ - - ' i S * u ^ ° r o - ' - * - * Hydrometeorologi (b) Rainfall (mm) co to co co co w to cal conditions (2) (c) Snowfall occurrence o o 25/-38/94 25/41 (d) Snowpack depth (cm) co cn A co M -» Runoff Event H FT C/3 C 3 3 o D. 1 O o -* ' o_ 5" °2. o' H. n o s a c 5 a. c Ui Vi JL n "» n rs TT CO C/3 n •O r+ O 3 cr n *1 VO vo I «H c s rt> I—I VO vo 00 Snow dapth (cm) . Dally rainfall (mm) Dally max/mln. temp. (C) 00 • i, I c 86 accumulation occurred during two main phases, following the two cP episodes. The second snowpack build-up was greater, with snowpack depth reaching its maximum in late February. Snowpack was relatively shallow during both cP episodes. The largest rainstorms occurred on 1 - 2 November (129 mm in two days), 2 - 3 December (108 mm in two days), and 28 February to 4 March (149 mm in five days). The early November rainfall occurred before any snowpack accumulation. The early December rainfall occurred soon after the first cP episode, with a minor snowpack on the ground. Rainfall may have been mixed with some snowfall, and snowpack accumulation occurred in the days following the major storm. The early March rainfall occurred when snowpack depth was at its seasonal maximum and air temperature was mild ~ prime rain-on-snow runoff generation conditions. Seventeen runoff events at the gully gauging stations were identified during the study period. Sixteen events occurred during the wet season, and one event occurred in mid-June due to a minor rainstorm. At both gully gauging stations and at Russell Creek Gauging Station, the early December rainstorm resulted in the maximum discharge of the study period, followed closely by peak discharge during the early March rain-on-snow runoff event. 4.1.2 Weather in Russell Creek Basin compared to Port Hardy Airport Hydrometeorological data presentation in Figure 4.1 begins on 10 September, 1993, to illustrate conditions immediately prior to the study period. During the 290-day period from 10 September, 1993, to 27 June, 1994, Russ2 weather data are missing for 132 days (46% of weather data display period). Weather conditions during periods of missing data were estimated by regression of air temperature and rainfall against Port Hardy Airport data for periods during which data were collected at Russ2. Regression results are shown in Figures 4.2 and 4.3. 87 t) Daily Minimum Temperatures R u s s 2 and Port Hardy Airport -10 -8 -6 -4 -2 0 2 4 6 8 10 12 Daily min. temp. (C), Pt. Hardy A — Regression^— 1:1 Line — Regression ^ — 1:1 Line Figure 4.2. Comparison of air temperatures at weather stations Russ2 and Port Hardy Airport, a) Daily minimum temperatures (R2=0.67). b) Daily maximum temperatures (R2=0.74). 89 Air temperature correlated moderately well between Russ2 and Port Hardy Airport. For daily minimum temperatures, R 2 was 0.67; for daily maximum temperatures, R 2 was 0.74. Comparison by seasons did not improve R 2 values. In general, minimum air temperatures in Russell Creek Basin were lower than at Port Hardy, especially on colder nights. Maximum air temperatures in Russell Creek Basin were generally lower than at Port Hardy in cold periods and higher than at Port Hardy in warm periods. These temperature differences are probably due to the lower elevation and the closer proximity to Johnstone Strait of the Port Hardy Airport weather station. Quantified temperature relations are: Daily minimum temperatures: TR2 = (0.92 x T P H ) - 2.2 Daily maximum temperatures: TR2 = (1.80 x TPH) - 7.0 where TR2 = Russ2 temp. (C), and T P H = Port Hardy Airport temp. (C). Rainfall regression was complicated by the probable occurrence of snowfall in Russell Creek Basin during some rainfalls at Port Hardy. Since snowfall was not recorded at Russ2 weather station, total precipitation amounts could not be compared. Therefore, the assumption was made that snowfall occurred at Russ2 during rainfall events at Port Hardy Airport when mean daily Russ2 air temperature was less than 1 C. Snowfall was also assumed to have fallen at Russ2 on days of snowfall at Port Hardy Airport. Estimated dates of snowfall occurrence are marked in Figure 4.1. Strength of the rainfall relation shown in Figure 4.3 was moderate (R2 = 0.78); in general, Russ2 received more rainfall than Port Hardy Airport. The relation may be non-linear, but for the purposes of this study the approximate linear relation suffices. Quantified rainfall relation is: R R 2 = 1.17 X R P H where RR2 = Russ2 rain (mm), and RPH = Port Hardy Airport rain (mm). 90 The relations between weather conditions at Port Hardy Airport and Russ2 weather station were considered strong enough to allow estimation of missing Russ2 conditions based on Port Hardy Airport data. 4.1.3 Study-period weather in perspective The long-term climate record at Port Hardy Airport allows the study-period weather conditions to be placed into perspective. Given the reasonably strong relation between Russ2 and Port Hardy Airport weather conditions, the comparison of study-period to long-term conditions at Port Hardy will be assumed to apply to Russell Creek Basin as well. Figure 4.4 shows average monthly precipitation and air temperature at Port Hardy Airport from 1961 to 1990, obtained from AES Climate Normals (Environment Canada 1993), as well as monthly values during the study period. The following observations can be made regarding the relation between study-period weather and long-term average weather conditions (ie. climate) at Port Hardy Airport, and the implications of these observations to Russell Creek Basin: 1) Autumn temperatures were about normal. September and October were unusually dry, followed by average precipitation in November and December. Therefore, the onset of the wet winter season was more pronounced than usual. 2) January was unusually warm and dry. Therefore, snowfall in the transient snowpack zone was probably much less than normal. 3) February was slightly cooler and wetter than normal. Therefore, snowfall in the transient snowpack zone was probably slightly greater than normal. 91 d. E 3 2> s Q S 4> E E £ Q. C o 2 Mean daily max. & min. temperature, 1993-94 Mean daily max. & min. temperature, 1961-90 Monthly precipitation, 1993-94 Monthly precipitation, 1961-90 Figure 4.4. Mean monthly precipitation and temperature at Port Hardy Airport during the study period compared to long-term average (1961 to 1990 Climate Normals). 92 4) The spring months were slightly warmer than normal with about average precipitation. Therefore, snowmelt in the transient snowpack zone was probably slightly more rapid than normal. Individual storm events during the study period can also be compared to the long-term average record. The two-year, 24-hour rainfall total at Port Hardy Airport, based on 1973 to 1990 rainfall records, was 86 mm. The maximum daily rainfall recorded at Port Hardy Airport during the study period was 76 mm on 2 November; next highest was 61 mm on 3 December. While these are large one-day rainfall totals, they are modest rainstorms by long-term Port Hardy standards. Return periods for multi-day rainfall totals were not available from AES. Only one two-day rainfall total during the study period exceeded the median one-day total —111 mm on 1 -2 November — so cutoff time in determining 24-hour rainfall totals was probably not a factor in underestimating the return period of the study-period storms. 4.1.4 Runoff events in Russell Creek compared to Tsitika River Peak discharge during independent runoff events was compared between WSC gauging stations on Russell Creek (08HF007) and on Tsitika River (08HF004), for the purpose of comparing study-period runoff in Russell Creek to the long-term records of Tsitika River. Russell Creek Gauging Station records begin in 1992, whereas Tsitika River records date from 1975. Thirty-six independent runoff events occurred at Russell Creek and Tsitika River gauging stations between July 1992 and June 1994 (ie. two complete runoff seasons). Comparison of corresponding maximum daily discharges during the runoff events is shown in Figure 4.5.' The strength of the relation between peak discharges at the two stations (R2=0.84) allows 93 o o CO o o LO j9A|y B>)H!SI ' ( S / E L U ) 86jeqos!a A||eQ CO CO co ro co • CN CO CO . • CO CD > H •a c a> CD 3 C O CD SP O C O C a> > ?t: o c s «S CD C U 1 o c o co 'C CS a, E o O 4> 2 s C/3 00 C '5b 3 o o oo co C > <L> o c 2 c cu -a c CD Q. <D T 3 C oo c • 3 T 3 cd cd CD L-c3 CD SP T3" 0 0 O II O N O i CD C 3 C N O N O N j > , 3 CO o s oo o E Q c CO CD C O C O 3 o o co o c 94 observations about study-period runoff at Tsitika River Gauging Station, compared to the long-term record, to be applied to study-period runoff at Russell Creek Gauging Station. The maximum daily discharge at Tsitika River Gauging Station from 1975 to 1994 was 617 m3/s in 1975. The greatest daily discharge at Tsitika River Gauging Station during the study period -- 445 m3/s --occurred on 3 December, due to the same rainstorm that generated maximum study-period flows at Russell Creek Gauging Station and at the gully gauging stations. The 3 December, 1993, Tsitika River discharge was the sixth-highest independent daily discharge in the 20-year record of the station. Therefore, this discharge has an estimated recurrence interval of four years (ie. exceeded five times in 20 years). The 2 -3 December rainfall at Port Hardy Airport was not notably large (less than median annual maximum). Rainfall at Russ2 during this storm lies only slightly above the regression line in Figure 4.3; therefore, Russell Creek Basin does not appear to have been subject to a locally intense rainfall cell. The notably large runoff must have been generated by a combination of reasonably heavy rain and prime snowmelt conditions. The second largest study-period runoff event at Tsitika River Gauging Station, as at the study gauging stations, was due to the early March rain-on-snow episode. Peak daily Tsitika River flow during this event was 265 m3/s ~ the fifteenth-highest daily flow on record. The Russell Creek discharge during this event lies considerably below the regression line in Figure 4.5; therefore, snowmelt in Russell Creek Basin must have been more significant than in Tsitika River Basin as a whole. Once again, anomalously heavy rain in Russell Creek Basin does not appear to have occurred; Russ2 rainfall during the early March event lies near the regression line in Figure 4.3 and the rain gauge was checked and cleared of ice immediately prior to the rainstorm (27 February), reducing the possibility of rainfall data having been missed. 95 4.2 Hillslope to gully-channel sediment transfers Hillslope to gully-channel sediment transfer activity at the monitored hillslope sediment-source sites is discussed below. Erosion / deposition pin exposure-length data may be found in Appendix A - l ; erosion / deposition between exposure-length measurements are presented in Table 4.3. 4.2.1 Sediment transfer activity at the monitored sites Site 1: Sidewall failure scar Activity on the failure scar consisted of accumulation of a thin veneer of material overlying the glacial-till failure plane, followed by removal of the thin veneer by shallow debris sliding. The veneer material may have been derived from ravel, sheetwash, and creep from the near-vertical soil exposure at the failure headwall, in addition to weathering of the till. During the monitoring period, five pins showed accumulation in the fall and early winter, followed by removal of surficial material during at least two separate winter-time events: pins # 1 and 2 were lost in a shallow (~ 70 mm depth) slide sometime between 6 December and 17 January, and pins # 3, 4 and 7 were lost in a slide of similar depth sometime between 17 January arid 13 May. Pin # 8 was subject to deposition during the late winter and gradual erosion in the late spring and summer. Two of the pins— # 5 and 6 --showed relatively little activity through the study period: Site 1 was apparently not subject to needle ice growth, since needle ice was not observed at this site on 6 December when it was observed at Site 2. Lack of needle ice growth was probably due to the impermeability of the underlying till, which prevented soil moisture migratiqn to the freezing front near the surface of the shallow sediment veneer overlying the till. 96 CO O CD n o (/>. o' 3 £0 3 Q . ? -O* Oo | ^  3 =>• Q . -i JU o W W 3 •g 3 ' CT c 3 . co 0} 3 Q . CO 3 Q . TJ 3 -_j. £ CD I to o </> 5" 3 3 3 o < CO o .7 » CD • • Q . Is * 1 Q . CD CD 3 j co r». CD a 3 cu CD o. ~-j o 3 3 o 3 CD 3 3 o < CD o. TJ 5' w ro "O 3 . O (0 (/) CD o ffl Nl U l M J J O) 4*. CO •73 o 5 H o T3 o .5 I I O) 2 2 2 2 • - » O ) 1 - i cn ro o 1 8 o" 3 2 2 D 3 3 ^ CO Co o a-o CO 3 5 CO , \ 1 cn ro ro co TJ 5' C/5 i-t-* rt rt <' rt CL to o TJ rt 1 I I _ i cn w co cn cn cn o ^ 4 ^ 1 o o 55 CO ro • N "NI ^ 0 0 0 ^ CO CO 3 1 3 o cn o < C O co o CD D CD p CO CO 2 I1 to ro , 9-1 3' CD 3 CO c o CD CO it ro. co e 3 tt rt s. B rt ( A rt tt "I H » cr rt" *». O s rt a rt o MB o s a. « •a o ( A o s tt o vr CO rt D. 3 rt s in O C rt rt ( A o •vl CO O) cn 4^  u to I I. I u M cn u O ^ N) s i z z z du J g w co ^ CO & M O M co o 10 ro - J co o ro TJ 5' cn co ^rt, rt OS o Q. rt S cr » s cn u M -k ro o co ro co ro a ID i co o ro cn CO co cn 70 5' =8= rt tn o in rt 3" cr tt B-?r o o CO CD . CO c 3 , £0 o CD CO CD Q . 3' . CD 3 ro cn cn CO ro cn co cn ro ro cn 31 5' CD CO c o CD CO CD a. 3' CD CO cn cn - A W 4k •>J Ji. CO CO CO ->l <=> CO tn o T) 3 ' o cn O CD O CO • CO c^1 p C O C O cn CO I' CO m , <=L\ 3 CD 3 i-» C/) c =3.1 CO o CD CO ro o j co s CO CD I 3 o 3 ro I co C O \0 Site 2 : Active channel bank At Site 2, the measurement pins were not all located in the eroding face; several were located in the depositional apron below. Further complicating data interpretation, the face / apron boundary shifted over time, so that pins could not be assigned constant "face" or "apron" attributes. However, pins #3,4 and 6 were located in a top row, and the others were located in a parallel bottom row, so these can be considered to approximate eroding-face and depositional-apron categories, respectively. Material at Site 2 was composed of sandy-matrix soil, with a long upslope soil moisture source; therefore, it was subject to needle ice growth. Long ice needles were present at the time of measurement on 6 December. On 17 January, net erosion had occurred at two of the three top-row pins, and net deposition had occurred at the four bottom-row pins which were still intact. Frost action between 6 November and 17 January may have been a main agent resulting in the transfer of material from the eroding face to the channel-margin apron. Between January and May, net erosion occurred at all pins, except one, probably indicating a downward shift in the face / apron boundary due to apron-toe erosion by gully-channel flow. The opposite trend ~ net deposition at all pins, except one ~ occurred between May and June. This represents apron recharge, possibly due to dry ravel on the eroding face. Site 3 : Sidewall failure scar Sediment transfer at Site 3 was measured only during one interval: 18 November until 6 December. During this period, the sediment-screen trap at the base of sidewall failure scar "F5" filled (approximately 0.3 m3) and overflowed. A frozen overflow waterfall was observed at the trap on 6 December. Particle-size of the trapped material ranged from fines to cobbles. Apparently, sediment transfer from the failure scar toward the gully-channel at its base occurred 98 by means of processes intermediate between sheetwash (evidence of surface runoff) and mass movement (unsorted sediments). Site 4 : LOD-jam sediment wedge Pins #1, 2 and 3 were located in fine sediments near the LOD jam at the snout of the sediment wedge; these are of primary interest to this discussion. Pins #4 and 5 were located toward the wedge apex in dominantly pebbly / cobbly sediments. The three pins in fine material showed variable minor sediment surface activity prior to 16 January. However, consistent deposition occurred between 16 January and 12 February, followed by consistent erosion before 20 March. The deposition may have been due to transport of sediment from channel-margin storage compartments — enriched by recent frost action ~ to the sediment wedge by the low to moderate gully-channel flows of late January to early February. Then much of the newly deposited sediment was probably eroded from the wedge by Runoff Event 11, generated by the major early-March rain-on-snow episode. Once again, variable minor activity occurred after 20 March. Site 5 : Cross-ditch bank All five pins were located in the eroding face of the channel bank. Greatest erosion occurred between January and May, and especially between March and May. However, response was variable between pins. Minor deposition and erosion occurred during other intervals. Site 5 may have been subject to needle ice development. However, this site was covered by snow on 12 February when needle ice was observed at Site 6. Drifting of snow off the road surface into the cross-ditch resulted in Site 5 being snow-covered for much of the winter; therefore, frost action may have been less significant than at the higher exposed bank sites (Sites 2 and 6). However, 99 absolute net erosion values over the course of the study period are similar at the three sites (maximum about 50 mm). The sediment-transfer similarity may be due to relatively shallow snowpack depths at all sites during the two cP anticyclonic cold spells. Site 6 : Road cut-bank Pins # 1, 2 and 3 were located in sandy-matrix soil receiving moisture from the'hillslope above and were subject to needle ice growth (observed on 12 February). Pins # 4, 5 and 6 were located in compact glacial till and were not subject to needle ice growth on 12 February. All pins were in eroding faces, just above the face / apron boundary. Maximum sediment transfer ~ erosion and deposition ~ occurred between January and March. Results were not clearly differentiated between the pins prone to needle ice growth and the others. Ambiguous, high activity rates may have been the result of various processes, including frost action and very small mass movements of moist, weathered bank material. After March, minor erosion was the dominant transfer activity, probably due mainly to dry ravel. 4.2.2 Hillslope sediment source activity and hydrometeorological conditions Erosion / deposition pin data are difficult to interpret. However, linking the patterns discussed in the previous section to the hydrometeorological patterns discussed in Section 4.1, the following summary observations can be made. The failure scar activity patterns were marked by gradual accumulation of thin sediment veneers over impermeable, glacial-till failure planes. Veneer material was then episodically discharged downslope to channel-margin storage compartments by means of very shallow (few cm), fluidised debris slides. The slide events on the monitored failure scars occurred during at least three separate events between November and May, probably during wet conditions when 100 gully channel discharge was high; if so, much fine sediment, delivered to the channel margin may have gone straight into down-gully transport. Depositional cones were present at the base of the failure scars year-round, but their relatively coarse texture supports the assumption of fine sediment throughput. Sandy-matrix banks with connectivity to hillslope soil moisture supply were subject to needle ice growth, though possibly to a lesser extent on banks where prolonged snow-cover occurred. Needle ice growth resulted in sediment transfers from eroding faces to storage aprons during and/or immediately following cold spells. Sediment was entrained from the apron toes during subsequent gully runoff events. Minor face-to-apron sediment transfers continued in the dry season, probably due to dry ravel. Minor runoff events following sediment-apron recharge periods probably transported sediment to the large sediment wedge, where at least some of it was deposited. Subsequent major runoff events entrained sediment from the wedge and from channel-margin storage compartments, as well as receiving direct sidewall contributions. 101 Chapter 5 Suspended Sediment Transport in Gullies Suspended sediment transport results from Gauging Stations B and C are presented in this chapter. Temporal patterns in the transport records are analysed with reference to hydrometeor-ological and sediment-supply conditions. Figure 5.1 displays discharge and suspended sediment concentration at Gauging Stations B and C during the study period. More detailed 10-day plots of discharge and SSC at the gully gauging stations may be found in Appendices A-2 (Station B) and A-3 (Station C). Table 5.1 summarises the portion of the study period during which data were successfully collected. Seventeen independent runoff events were identified at Gauging Stations B and C. Although the same set of runoff events occurred at both stations, hydrologic response differed between the stations. Most of the recorded study-period suspended sediment transport at Gauging Stations B and C occurred during the 17 runoff events. Table 5.2 summarises hydrologic and sedimentologic characteristics of the 17 runoff events. Analysis of temporal suspended sediment transport patterns at the gully gauging stations deals only with runoff-event transport. The occurrence of a debris torrent in Gully B shortly after the study period provided the opportunity to make a rudimentary comparison between fluvial and debris-torrent transport of fine sediment in the gully. 102 Q (1/6) I l l l l l f f I I I miff Q (l/s) f a O cr d3 rt s o. CO C CO XI rt CJ C L o> C L CO rt C L 3 rt 3 O o 3 0 CD 3 r-t-' >-l P rt-o ° •3' 1 5' 0Q o 3 co Cd C L O 00 n rt c l rt VO C 3 rt VO VO S S C (mg/1) S S C (mg/1) © Table 5.1. Summary of data collection at Gauging Stations B and C Period Duration (days) Fraction ot study period (%) Stn. B Stn. C Stn. B Stn. C Study period: 30 Sep. /93 - 27 Jun. /94 270 270 100% 100% -Discharge data collected 229 211 85% 78% Discharge and SSC data collected 204 176 76% 65% 5.1 Hydrological comparison of Gullies B and C Study Gullies B and C were selected to be subject to the same, synoptic-scale hydrometeorological conditions, hence, their close proximity and similar aspect and elevation. However, the gullies were also chosen to represent different gully types, which results in different hydrological responses. In particular, vegetation cover plays an important role in energy and water transfers at the local scale. Hydrological characteristics of the study-period runoff events were compared between Gauging Stations B and C to assess similarity of relative event magnitudes at each station and to assess overall differences in hydrological responses between stations. Absolute peak discharges and total runoff volumes during the study-period runoff events are compared between the two stations in Figure 5.2. Specific discharges and runoff volumes are compared in Figure 5.3. Comparison was possible for 13 of the 17 study-period runoff events in which flow data were successfully collected at both stations. Figure 5.2 (a) shows that a moderately strong, positive relation exists between peak discharge at the two stations. Therefore, the relative importance of a given runoff-event peak discharge can be assumed to be similar at both stations. Figure 5.2 (b) shows that event runoff 104 o s <S CQ e a « Co OA s o > v fc o c 3 ]_ c~ o E E 3 C/j IT) •O a H iment Transport Total Transported Mass (kq) c c 2= \f> o o> m' n V ci 2 iment Transport Total Transported Mass (kq) ct c 55 •r- ^ CO £: co o> CD " <» -r- TJ- CM T - TJ- 2^ TJ- JC) IO 2 Suspended Sec Maximum 15-min. Concentration (mq/l) u c 25 CM *- * " T -5 >*- *5 "3 -> "> c m S S S IO O OI ID T ; T ; n/a Suspended Sec Maximum 15-min. Concentration (mq/l) m c 25 "1 CM CM O CO co ° w o V » TJ- CO co CM » CO <o r~' CM to $ CD 2 at Gauging Stations Total Runoff Volume (x 1000 m3) I Stn. C T- CM CO CM Y - r - O CO i— LO IO Tj- CO T J CT> 1 Y— CO Y-- CM* O o at Gauging Stations Total Runoff Volume (x 1000 m3) m c 25 T - N m O •2 Y - T - CM O TJ- CO CD O) O to CO Y - TJ- Y - CO CO T - OJ CM CO CT> CO T - CO Y - T- Y -OS Response; 5-min. rqe (l/s) o cf 55 CO CO Y— 0> r W N t -CD CM CO TJ- CM "> cn T- CN ^ co (o Y— Runoff Max. 1 Discha co er 25 _ i - co T -2 CO IO CM CM CM Y - OT CM CO CM 1^ O T- CO N io CO T - TJ- CM CM T - CM OI d l N CO N COT O r - C O ° CO CM Y— CM CO jns (1) Type (2) T - T - CM CM CM CM CM CM CO CM CM CM CO CM CM CO tion Conditit Pre-event snowpack (cm) O O o o *~ T~ i i o o ° ° 2 § C M S A A A o 'ent Genera Temp, range + 18/+5 +19/+1 +6/0 +7 / +1 +5/-1 +8/0 +3/-1 +0 / U +12/+5 +9/+1 +13/-2 + 13/-1 +25 / +3 +y / +5 +7/+1 +23 / +5 +21 / +5 Runoff E\ Total rainfall (mm) 2J o> CO TJ- 5-! •* co RO CO CM TJ- 2 CM _ CO Y - _ J CM ° Y - TJ- <"> CM CO Date (1993-94) 22 - 24 Oct. 02 - 04 Nov. 14-16 Nov. 20 - 22 Nov. 29 - 30 Nov. 02 - 04 Dec. 09 -11 Dec. 28 Dec. - 05 Jan. 12-14 Jan. 21 -24 Jan. 27 Feb. - 04 Mar. 12-14 Mar. 26 Mar. - 01 Apr. 01 - 03 Apr. 11 -12 Apr. 14- 18 Apr. 12-13 Jun. c LU T- CM CO TT m to s co cn ° •— C\J CO l O CD Season and Sub-Season 13 5 " 5 -"S — 5 = Dry o o Q. I i CD fe.l ra o *I O CL — CO S | eu i 3 co TJ CO o •a co co > f 5 cu CU c O | co W g co CO co II s l l , 2 cu a . TJ C 8 J" •- — "O Sz CO CO CO o DC CC or Q; Y- CM CO CO ro TJ TJ CM £ •& £ 3 fc cu y CO CO ™ CU CU TJ O Q . CU CO 3 E "2 3 8 3 8 w S f S §• fe 5 3 ^ ro eo £ 2 cu 2 5 CO o 105 a) Peak Discharge 1000 o s co 100 S 10 10 11 2 ' " 1 4 ; t i t 3 - " 1 3 : 4 •16 - + — — i — i — i — i i 1 i - i 1 — i — i — f — i i i 100 Peak Q (l/s), Stn. B 1000 b) 100 o £ co S 1 O c or 0.1 Runoff Volume 1 3 ¥ 6 2 J" 1 5 -1 2 1 4 * 1 6 • 1 0 -i i — » 1 7 i i i j 10 Runoff Vol. (x 1000 m3), Stn. B . . i i . -i - i 1—i—i—i i i 100 Figure 5.2. Hydrological comparison of Gullies B and C. Comparison of a) peak discharges, and b) runoff volumes, during the 13 study-period runoff events for which flow data are available from both stations. 106 b ) 10 o d to CO E 1 4: •5 ° - 1 + ex CO 0.01 0.1 Speci f ic Event Runoff Vo lume 14 3 « , 6 H 1 1—I—I—I—H •—I ^ 1 Spec. Vol. (x 1,000 m3/ha), Stn. B 10 Figure 5.3. Areally-normalised hydrological comparison of Gullies B and C. Comparison of a) specific peak discharges, and b) specific runoff volumes, during the 13 study-period runoff events for which flow data are available at both stations. 107 volumes are also positively, but more weakly, related. Gauging Station B experienced relatively high runoff volume during runoff events 10 and 16. Melting of a shallow snowpack was a factor in the generation of both runoff events. Since snowpack depth was monitored at clearcut sites, snow may not have been present in the forest prior to these events, resulting in total runoff at Station C below the normal relation. Figure 5.3 shows that during the largest runoff events, peak discharges and event runoff volumes at the two gauging stations were approximately proportional to drainage area. During smaller runoff events, however, specific peak discharges and event runoff volumes at Gauging Station C were considerably lower than at Station B. Apparently, lack of forest cover in Gully B had considerable hydrological effect during small runoff events, with diminishing influence during increasingly large events. 5.2 Importance of runoff events in transporting suspended sediment Table 5.3 summarises total recorded runoff volume and suspended sediment transport during the study period and during study-period runoff events at Gauging Stations B and C. Suspended sediment transport during runoff events accounted for at least 90% of the total recorded suspended sediment transport at the gully gauging stations during the study period. In large part, transport occurred during runoff events because these accounted for about three-quarters of total runoff, despite occupying only about one-fifth to one-quarter of the data collection period. In any case, focussing attention on runoff-event suspended sediment transport seems appropriate. Very low suspended sediment concentrations below the lower limit of SSC resolution (1 mg/1) may have resulted in underestimation of total transport, especially during the relatively long intervals between runoff events. To check this possible source of error, suspended sediment 108 . Importance of runoff events in transporting suspended sediment at Gauging Stations B and C 1 Period / Item I Stn. B I Stn. C Discharge and SSC data collection period: Total runoff volume (x 1,000 m3) 520 49.4 Total suspended sediment transport (kg) 2921 403 Runoff events with recorded discharge and SSC data: Number of runoff events with recorded discharge and SSC data: 15 12 Duration of runoff events (days): 52.0 33.8 Duration of runoff events as fraction of discharge and SSC data collection period (%): 25% 19% Total runoff volume (x 1,000 m3): 388 38.0 Runoff volume during runoff events as fraction of data collection period total: 75% 77% Total suspended sediment transport (kg): 2643 379 Suspended sediment transport during runoff events as fraction of data collection period total: 90% 94% cn-633.wb1 transport was recalculated, assuming a minimum SSC value of 1 mg/1 during periods in which recorded SSC was less than 1 mg/1. Under this condition, total suspended sediment transport values increased by 237 kg at Station B and by 14.8 kg at Station C, representing 8% and 4%, respectively, of original total transport, values. However, fractional changes in relative runoff-event contributions to study-period transport totals were less, because low SSC occurred both during and between runoff events. Assuming minimum SSC of 1 mg/1, runoff-event suspended sediment transport still comprised 89% and 92% of study-period total transport at Stations B and C, respectively. 5.3 Seasonal scale patterns The study period covered the late part of the 1993 dry season, the entire 1993-94 wet season, and the early part of the 1994 dry season. Wet and dry seasons were clearly reflected in the temporal pattern of suspended sediment output from Gullies B and C because runoff events occurred almost entirely during the wet season. Suspended sediment transported by individual runoff events during the study period was not consistent. Figure 5.4 shows that relations between event peak discharge and peak SSC are weak at both Gauging Stations B and C. More significantly, relations between event runoff volume and suspended sediment transport are also weak, despite the spuriously inflated nature of the correlation (Figure 5.5). Lack of strong relations between total runoff volume and suspended sediment transport are most likely due to temporal variability in sediment supply. Runoff volume and suspended sediment transport are plotted event-by-event in Figure 5.6 to allow examination of temporal patterns in the runoff-transport relations. The often quoted pattern of suspended sediment transport decline with successive runoff events throughout the runoff season is not evident. 110 a) 1000 1 1 0 0 E_ O co co CO CO o. 10 10 Gauging Station B MIS" • 2 . .«6 .. Mil ' '< '• -16< :13 v , io : 8 J " J -t 1 1 1 1—I i i i M9 -4- - i 1 — i — i i i 100 Peak Q (l/s) 1000 b) 1000 e i o o CO E, O co to 3£ CO CD Q. 10 Gauging Station C 10 100 Peak Q (l/s) • 1 ) - J - v r ; t T t ». . . .». . . , . . . • . . , .» . , . : • 13 - - : i .i !...]...! i j.... ...— , 1 \-f f^4^--\ f~ ::::::::::::::!::::: kilj;yii;:::;::::::i::::::r:4:4::i::l:^: ! 1 mlO i H 1—1 M i l ] H* —1 i i i i i i i i—i—i i i M i 1000 Figure 5.4. Comparison of peak suspended sediment concentration to peak discharge during study-period runoff events at a) Gauging Station B, b) Gauging Station C. I l l a) 10000 "55 Q.1000 co c CO 100 + . <z cu E X) CD CO T J cu T J I 10 co CO Gauging Station B ::::::::::::::::::::[:::: : ' ; : ' - j "11 _ _ L... ! i I i : : : ; i i • ~ f : : 2 ; M I S • 13 t '1"f" - T : : : : : : : : : : : : : : : : : : : i : : : : : : i i : : : : : : : : : : : : : : : : : : ! : : : : : : £ " ' - ^ -s-14 - , ' 6 :::£:::::::•::::::•::::;:: | } j h H - i " ! • f - - • 1 0 • 8 " : : : : : : : : : : : : : : : : : : : l : : : : : • 3 . . ; . . . j . . J . . . ::::::;:::::::?:::::l:::Si:iS ?::!::::::::::::::::::::::«:4 : ::::::::::::!::::::;::::::: 1 : : . " 9 1 H 1 1 1 1 1 | 1 —1 1 1—1- i i i 10 Runoff Vol. (x 1000 m3) 100 b) 1000.0 c CD E T J CD CO T J CD T J C CD C L CO D CO 100.0 10.0 1.0 0.1 Gauging Station C • ::t:::::|::::|::::!::;::r3:: "•13 • 11 : . j - - 6 . ~ : - « 1 2 - j • 14 • f . . - . i - . . . { - " i - - - j - 4 4 ; - --™1 mi ::::::::::::::::r:::::4::::::::^  3,10 • W4 i i - i — i — i — i — i - i - i — 10 Runoff Vol. (x 1000 m3) 100 : , , , I Figure 5.5. Comparison of suspended sediment transport to runoff volume during study-period runoff events at a) Gauging Station B, b) Gauging Station C; 112 a) 120 Gauging Station B + 1000 7 8 9 10 11 12 13 14 15 16 17 Runoff Event I | Runoff Vol. | Susp. Sed. Transp. 1200 800 600 8. c CO E to 4400 -s 200 •o c CD CL to b) 16 14 <| 12 o 8 10 CD E 4: o c 2 + Gauging Station C n 1 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Runoff Event | | Runoff Vol. | Susp. Sed. Transp. 160 Figure 5.6. Event-by-event display of runoff volume and suspended sediment transport during study-period runoff events at a) Gauging Station B, b) Gauging Station C. 113 However, two other possible patterns are evident. Firstly, sediment supply may be limited below a threshold discharge, only above which does significant suspended sediment transport occur. Secondly, sediment supply may undergo sub-seasonal cycles of depletion and replenishment. Runoff event data are examined for the occurrence of these patterns in the following section. 5.4 Sub-seasonal scale patterns In Figure 5.7, suspended sediment transport during runoff events at Gauging Stations B and C does not show clear dependence upon event peak discharge, although a threshold discharge is evident at Station B. However, if sub-seasonal occurrence of runoff events is considered, clearer patterns emerge. 5.4.1 Gauging Station B At Gauging Station B, a suspended.sediment transport threshold appears to occur at about 270 l/s peak discharge. Above 270 l/s, suspended sediment transport increases dramatically, regardless of sub-season. The peak-discharge threshold probably results from high flows accessing channel-margin sediment supply that is only infrequently exposed to flow. Sediment supply at elevations accessible by sub-threshold flows is frequently depleted by runoff events, but when flows gain access to the higher-elevation sediment supply, transport suddenly increases. Figure 5.8 shows that suspended sediment transport during large runoff events is well related to the runoff volume that occurs while discharge is greater than 270 l/s. RunoffEvent 7 probably plots low for transport because it immediately followed the larger Event 6 with little time between for supply recharge. Below 270 l/s peak discharge, transport-discharge relations appear to be sub-seasonally controlled (Figure 5.9 (a)). In Sub-season I, suspended sediment transport was relatively low in 114 a) 10000 CD &1000 in c s « 100 E s 10 & co CO 10 Gauging Station B Threshold Q ~ 270 l/s 13,15 -16 •14 |9 * H 1 1 1 1—I | | 11 « . . . 06 100 Peak Discharge (l/s) H 1 1 1—I-1000 b), 1000.0 & 100.0 CO c CO c CD E TJ CD CO "2 T J C CO n CO 10.0 1.0 J 0.1 Gauging Station C • 13 • 11 • 6 • 12 • 14 Ml .10 , , , 3 4 - i 1—i t i i i | ' I I I I I | 10 100 Peak Discharge (l/s) -+.—i—i i i i i 1000 Figure 5.7. Comparison of suspended sediment transport to peak discharge during study-period runoff events at a) Gauging Station B, b) Gauging Station C. 115 Gauging Station B 10000 4 • , •—. . \. . - — 10 -I i i—\—i i i i i i i -i—I—i i j j | I 1 10 100 Runoff Vol. while Q>270 (x 1000 m3) Figure 5.8. Comparison of suspended sediment transport to runoff volume above threshold peak discharge (270 l/s) during large runoff events at Gauging Station B. sub-threshold Events 3 and 4. Two runoff events had already occurred since the onset of the wet season, during at least one of which peak discharge exceeded threshold. Therefore, dry season accumulations of sediment along the channel margin would have likely been depleted prior to Events 3 and 4. Further supporting the theory of ongoing sediment supply depletion, suspended sediment transport was less during Event 4 than during Event 3, despite higher peak discharge and runoff volume during Event 4. In Sub-season II, suspended sediment transport increased relative to Events 3 and 4, perhaps due to replenishment of channel-margin sediment supply by frost action. During Event 5, suspended sediment transport was similar to Event 3 and 4 transport, despite the much lower flow. After the large runoff events 6 and 7, transport by given peak discharge appears reduced relative to early sub-season Event 5, but still greater thanduring Sub-season I. 116 a) Gaug ing Station B 10000 &1000 CO c CO CO T J CU "D C 8. CO D CO 10 10 :Thrcshol«lQ~270l/s • H 1 1 1 1—I I I 100 Peak Discharge (l/s) 1000 b) Gauging Station B 600 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Runoff Event 1200 1000 2 r o CL 4-800 600 + 400 + 200 TJ CD CO T J CD TJ C 8. CO CO I | Discharge | Susp. Sed. Transp. Figure 5.9. Influence of peak discharge on suspended sediment transport during runoff events at Gauging Station B. a) Transport versus peak discharge, sub-seasonal trends identified in sub-threshold events, b) Event-by-event display of transport and peak discharge, threshold discharge identified. 117 In Sub-season III, suspended sediment transport by given peak discharge increased yet again, even though Events 12 through 16 followed the large Event 11 which transported the most sediment of any during the study period. Sediment supply must have been replenished on an ongoing basis throughout Sub-season III, presumably by frost action and snowmelt. Frost action probably occurred not only during the cP anticyclone early in Sub-season III, but also on cold clear nights under subtropical anticyclonic conditions, when nighttime temperatures often dipped below 0 C. Snowmelt rivulets and small avalanches of wet snow with embedded sediment were observed to transport sediment to the gully channel margin. Frost action and snowmelt processes both operated throughout Sub-season III. In Figure 5.9 (b), the combined effects of peak-discharge threshold and sub-seasonal control on discharge-transport relations can be seen. Runoff Events 6 and 11 ~ with the highest peak discharges — dominated suspended sediment transport during the study period (27% and 38% of study-period total, respectively). Four runoff events which barely surpassed 270 l/s each transported 100 to 200 kg of suspended sediment, or 3% to 6% of study-period total per event. Apparent sub-seasonal supply depletion during Events 3 and 4, and 8 through 10, may have been partly due to sub-threshold peak discharges during these events. The meaningful comparison is between the aforementioned events and the Sub-season III sub-threshold runoff events: 12, 14 and 16. During the latter three events, sediment transport ~ although low compared to threshold-exceeding event transport ~ was consistently greater than during the comparable Sub-season I-and II events. The occurrence of Event 11, with the greatest flow duration in excess of 270 l/s, following a cold spell, and at the beginning of the snowmelt period complicates disentanglement of sub-seasonal patterns from peak-discharge threshold effects. Overall, however, discharge characteristcs appear to more strongly control suspended sediment transport in Gully B than do sediment supply conditions. Sub-seasonal patterns in 118 sediment supply conditions produced clear patterns in suspended sediment transport during small to intermediate runoff events, but transport during such events was at least one order of magnitude less than during the largest events of the study-period. Although abundant exposed hillslope sediment sources occur within Gully B, sub-seasonal patterns of sediment transfer from the sources to channel-margin storage compartments may be of minor importance because of storage buffering. 5.4.2 Gaug ing Station C Suspended sediment transport at Gauging Station C does not appear to be controlled by a peak-discharge threshold. However, sub-seasonal patterns in discharge-transport relations were evident during the study period (Figure 5.10). In Sub-season I, Event T resulted in the largest suspended sediment transport, despite relatively low peak discharge. During Events 2 through 4, all with peak discharges exceeding Event 1 peak discharge, a relatively strong discharge-transport relation existed, with transport much less than Event 1. Event 1 runoff probably transported the most sediment because of dry season accumulation in and adjacent to the gully channel. During subsequent runoff events, once the summertime accumulation had been removed, a lower, stable rate of sediment transport was reached. Data from Gauging Station C during much of Sub-season II are missing. Suspended sediment transport during Event 6 was similar to that of Event 11; transport during Event 10 plots slightly above the depleted-supply runoff events of Sub-season I. The relations of transport between these events at Station C are similar to corresponding inter-event relations at Station B. \ In Sub-season III at Station C, as at Station B, a relatively strong discharge-transport relation exists with the greatest sediment transport of the three sub-seasons. Much different from Station B, Event 13 plots anomalously high, with the greatest suspended sediment transport of the 119 a) Gauging Station C 1000.0 & 100.0 c c CD E co "S T J C & V) 3 to 10.0 1.0 0.1 " • 1 3 :::!::J::!:!:i:::::::::::::l::~ : * i -j::::;:::: : 1 V - • • M 1 0 *0<i ' Z 4 E 1—1 1 1 1 I 1 1 V—1 1 10 100 Peak Discharge (l/s) 1000 b) 160 140 120 CD CD k_ CO SZ <J CO CD 0_ 100 Gauging Station C 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Runoff Event | | Discharge | Susp. Sed. Transp. Figure 5.10. Influence of peak discharge on suspended sediment transport during runoff events at Gauging Station C. a) Transport versus peak discharge, sub-seasonal trends identified, b) Event-by-event display of transport and peak discharge. 120 study period -- 37% of study-period total -- despite modest peak discharge. The radiative snowmelt which generated Event 13 must have resulted in prolonged sediment transfer into the gully channel over the course of the multi-day event. During reconnaissance of Gully C at the end of the study period, I did not observe any sidewall failure scars; therefore, the prolonged sediment transport during RunoffEvent 13 was not due to a fresh sidewall source. Sub-seasonal patterns in sediment supply conditions appear have be more important in Gully C than in Gully B, despite the less abundant hillslope sediment sources and channel-margin storage compartments in Gully C, probably due to the minimal buffering in the system. Dry season processes — such as dry ravel and organic litter accumulation ~ and snowmelt-related processes ~ such as minor avalanches and sheetwash transporting organic debris — were probably the main sediment supply replenishment agents in Gully C. The suspended sediment load at Gauging Station C may have been more dominantly organic than at Station B. Indeed, the similarity of turbidity-SSC relations at the two stations (Figure 3.6), despite the coarser sediment source texture in Gully C (Table 2.1), is probably due to counteractive effect of higher organics content in Gully C suspended load. 5.4.3 Summary Figure 5.11 shows suspended sediment transport versus runoff volume at Gauging Stations B and C separately. Runoff events at Station B are categorised as: 1) Sub-seasons I and II with peak discharge above threshold and all Sub-season III, and 2) Sub-seasons I and II with peak discharge below threshold. Runoff events at Station C are categorised as: 1) Sub-seasons I and II first-events and all Sub-season III, and 2) Sub-seasons I and II, depleted. The categorised data from the two gully gauging stations in Figure 5.11 are plotted in a single graph in Figure 5.12. Interestingly, the categories defined in Figure 5.11 correspond at 121 a) Gauging Station B 10000 9-1000 c ro E 100 to T J C 10 to JML. ••-t--< 4 . .«10,. : . . B t 8 10 RunoffVoL (x1000 m3) 1^ 100 b) Gauging Station C 1000.0 100.0 c cu E T J CD to T J CD T J C 8. in to 10.0 1.0 0.1 • 12, I * l l - ' •3,10 a.S-• W4 < 2 . S . ^ ' H 1 1 1-^ —I I I —t I t I—I—h+ 10 Runoff Vol. (x1000 m3) 100 Figure 5.11. Comparison of suspended sediment transport to runoff volume during runoff events categorised by peak-discharge and sub-season, a) Gauging Station B. b) Gauging Station C. 122 both gauging stations. The categories in Figure 5.12 are referred to more generally as "replenished supply" and "depleted supply". Median relations within each category are: 1) Replenished supply: T = 9 . 0 V ° 9 5 2) Depleted supply: T = 1.1 V ° 6 9 where T = suspended sediment transport (kg) and V = runoff volume (x 1000 m3), for individual independent runoff events. Slope of the replenished supply relation in Figure 5.12 is 0.95 — nearly unity. Slope = 1 can be defined as constant event-averaged SSC, since suspended sediment transport is the product of runoff magnitude and mean SSC. In other words, average SSC is independent of runoff event magnitude when slope =1. The depleted supply relation, with slope much less than unity (0.69), means that during increasingly large runoff events additional suspended sediment transport does not match additional water discharge, resulting in increasingly dilute flows (ie. lower mean SSC). In the two gullies studied, slope = 1 in transport-runoff relations may define an upper limit of sediment supply replenishment. In Figure 5.13, runoff-transport relations are normalised by drainage area and channel length for more general comparison. Channel length was chosen as the sediment-supply normalising factor assuming that sediment is largely entrained along channel margins. Median areally-normalised relations are: 1) Replenished supply: T L = 90 V A 1 2 4 2) Depleted sediment supply: T L = 7.0 V A ° 8 2 where T L = suspended sediment transport per unit channel length (kg/km), V A = runoff volume per unit area (x 1000 m3/ha), for individual independent runoff events. 124 Slopes of the areally-normalised sediment supply-condition relations bracket unity. In other words, the replenished supply condition may be defined as transport per unit length of channel increasing more rapidly than runoff per unit drainage area in increasingly large events. Conversely, the depleted supply condition may be defined as unit transport increasing less rapidly than unit runoff. Resultant trends in SSC cannot easily be referred to in the normalised case because normalisation dimensions differed for transport and runoff. The degree to which these generalisations are transferable cannot be determined without future studies. 5.5 Synopt ic scale patterns Characteristics of suspended sediment transport within the 17 study-period runoff events are summarised in Table 5.4. Seven runoff-event suspended sediment transport records in which peak SSC exceeded 20 mg/1 were selected for synoptic-scale study. Smoothed chronological plots of discharge versus SSC (Q-SSC plots) during runoff events provide greater understanding of temporal relations between flow and sediment supply. Observed Q-SSC classifications in Table 5.4 refer to the literature review provided in Table 1.3. 5.5.1 RunoffEvent 1 at Gauging Station C: The 22 - 24 October runoff event was generated by 41 mm of rainfall and no snow-melt. This was the first runoff event of the 1993-94 wet season. At Gauging Station C, the event hydrograph was double-peaked (Figure 5.14), and antecedent flow had been negligible since the start of the study period (22 days) and probably for at least six weeks, based on hydrometeorological data starting 10 September (Figure 4.1). A short pulse of suspended sediment transport occurred on the rising limb of each hydrograph peak and on the falling limb of Hydrograph Peak 1. 126 • Table 5.4. Synoptic-scale characteristics of suspended sediment transport during runoff events at the gully gauging stations Gauging Station B Runof ' Event Suspended Sediment Transport Event Date Generation Duration (1) Max. 15-min. Duration of S S C S S C exceedence duration Chronological # (1993-94) Type (hr.) S S C (mg/l) exceedence (hr.) as % event duration Q vs. S S C relation •* 100 mg/l 20 mg/l 100 mg/l 20 mg/l type (3) 1 M 1 M M M M M M 2 01 - 04 Nov. 1 84 171 0.25 3.00 0.3 3.6 S-2 & S-6 3 1 4 - 1 7 Nov. 2 72 4 0.00 0.00 0.0 0.0 -4 1 9 - 2 2 Nov. 2 72 1 0.00 0.00 0.0 0.0 5 29 - 30 Nov. 2 36 10 0.00 0.00 0.0 0.0 6 02 - 04 Dec. 2 54 102 0.25 12.25 0.5 22.7 M-2 7 09 -11 Dec. 2 48 22 0.00 0.25 0.0 0.5 S-2 8 28 Dec. - 05 Jan. 2 204 3 0.00 0.00 o.o y 0.0 _ 9 1 2 - 1 4 Jan. 3 54 1 0.00 0.00 0.0 0.0 -10 21 - 24 Jan. 2 78 5 0.00 0.00 o.o 0.0 -11 27 Feb. - 04 Mar. 2 126 64 0.00 10.25 0.0 8.1 M-2 & M-4 12 1 2 - 1 4 Mar. 2 60 8 0.00 0.00 0.0 0.0 -13 26 Mar. - 01 Apr. 3 144 26 0.00 0.75 0.0 0.5 M-3 14 01 - 03 Apr. 2 48 8 0.00 0.00 0.0 0.0 ~ 15 1 0 - 1 2 Apr. 2 48 268 0.25 0.50 0.5 1.0 S-3 16 14 - 1 9 Apr. 3 120 7 0.00 0.00 0.0 0.0 17 1 2 - 1 3 Jun. 1 36 M M M M M -cn-633.wt>1 Gaug ing Station C Runoff Event Event # Date (1993-94) Generation Type Duration (2) (hr.) Suspended Sediment Transport Max. 15-min. S S C (mg/l) Duration of S S C exceedence (hr.) 100 mg/l | 20 mg/l S S C exceedence duration as % event duration 100 mg/l | 20 mg/l Chronological Q vs. S S C relation 1 2 3 4 22 - 24 Oct. 02 - 04 Nov. 1 4 - 1 6 Nov. 20 - 22 Nov. 1 1 2 2 72 60 48 60 243 10 11 1 0.50 0.00 0.00 0.00 1.50 0.00 0.00 0.00 0.7 0.0 0.0 0.0 2.1 0.0 0.0u 0.0 M-2 & M-6 5 6 7 8 9 10 29 - 30 Nov. 02 - 04 Dec. M M M 21 - 24 Jan. 2 2 2 2 3 2 36 60 M M M 72 n/a 57 M M M 3 n/a 0.00 M M M 0.00 n/a 2.50 M M M 0.00 n/a 0.0 M M M 0.0 n/a 4.2 M M M 0.0 S-2 11 12 13 14 I yJ 16 27 Feb. - 04 Mar. 1 2 - 1 4 Mar. 26 Mar. - 01 Apr. 01 - 03 Apr. 11 - 1 2 Apr. 1 4 - 1 8 Apr. 2 2 3 2 2 3 126 54. 144 60 - 4 8 -108 35 18 52 16 - M -M 0.00 0.00 0.00 0.00 - M M 1.75 0.00 100.25 0.00 — M — M 0.0 0.0 0.0 0.0 0.0 0.0 1.4 0.0 69.6 0.0 0.0 0.0 M-3 M-1 17 1 2 - 13 Jun. 18 n/a n/a n/a n/a n/a cn-633.wb1 Notes: 1. 2. 3. 4. Runoff event durations defined from 10-day discharge plots, Appendix A-2. Runoff event durations defined from 10-day discharge plots, Appendix A-3.» Chronological Q vs. S S C relation types listed in Table 1.3. " M " = missing data to a) Discharge (Q) and Suspended Sediment Concentration (SSC) versus Time SSC Peak lb ' ' i T I 1 1 i j • j— 1— 1 j (-1 18:00 00:00 06:00 12:00 18:00 00:00 06:00 12:00 18:00 0000 0600 21 Oct. 22 Oct. 2 3 0 c | 2 4 0 c , Time (PST) Q SSC b) Temporal Frequency Distribution of SSC Exceedence = 2.0 D j l . O i ! 0.5 ! oo 10 100 SSC (mg/l) . 1000 c) Chronological Q versus SSC 125 100 U A w SSC Peak lb 1 75 - \ | co 50 CO 25 0 Y SSC Peak la 10 Q (l/s) 15 20 Figure 5.14. Discharge (Q) and suspended sediment concentration (SSC) at Gauging Station C during Runoff Event 1, 21 - 24 October, 1993. a) Q and SSC versus time, b) Temporal frequency distribution of SSC exceedence during runoff event, c) Chronological Q versus SSC plot. 128 Maximum SSC on the rising limb of Hydrograph Peak 1 ~ labelled SSC Peak la -- was about 10 times greater than SSC Peak 2 on the rising limb of Hydrograph Peak 2, despite a peak-discharge ratio between the hydrograph peaks of less than 2 : 1 . Even this minor discharge difference does not explain maximum SSC difference since the rising-limb flows during SSC Peak la were less than the peak flow of Hydrograph Peak 2; in other words, the two rising-limb SSC peaks occurred under similar flow conditions. Most likely, the high SSC value on the rising limb of Hydrograph Peak 1 was due to transport of easily mobilised sediment that had accumulated in and immediately adjacent to the channel during the dry season. Depletion of this sediment supply had occurred prior to Hydrograph Peak 2, resulting in the much lower peak SSC. SSC Peak lb occurred on the falling limb of Hydrograph Peak 1. This was the highest SSC recorded at Gauging Station C during the entire study period. Because it occurred on a falling hydrograph limb, one may infer that this transport pulse was probably due to a random, discrete event, such as a bank collapse or an LOD jam disruption. One cannot determine whether the occurrence of this sediment-supply event during the first runoff event of the wet season was a coincidence, but the first thorough soil wetting after the dry season makes bank collapse a credible hypothesis. The chronological Q-SSC pattern plotted in Figure 5.14c consists of a clockwise loop, representing the rising-limb entrainment of accumulated material during SSC Peak 1, followed by a counter-clockwise loop, representing the random, falling-limb sediment input during SSC Peak lb. SSC Peak 2 barely shows up in the plot. 5.5.2. RunoffEvent 2 at Gauging Station B: The 2-3 November runoff event was generated by 129 mm of rainfall and no snow-melt. At Gauging Station B, the event hydrograph was single-peaked, with steep rising limb and gentler 129 falling limb (Figure 5.15). Peak discharge exceeded the "replenished supply" discharge threshold of 270 l/s for about 3.3 hours. Two main suspended sediment transport pulses occurred during the runoff event. The highest SSC — labelled SSC Peak 1 — occurred in conjunction with the hydrograph peak. SSC Peak 2 actually consists of several SSC peaks which occurred in rapid succession on the falling limb of the hydrograph. During SSC Peak 1, maximum discharge almost coincided with peak SSC, and the 270 l/s discharge threshold coincided with rapid SSC increase on the hydrograph rising limb, indicative of the replenished sediment supply which can be accessed by high flows. SSC was greater at the onset (40 mg/l) than at the cessation (5 mg/l) of threshold discharge, indicating that suspendible sediment was available to even sub-threshold rising-limb flows, and that supply depletion occurred even in the upper channel-margin sediment storage compartments which are only infrequently accessed by high flows. The series of falling-limb SSC pulses (SSC Peak 2) was probably due to a random sediment-supply or sediment-release event, such as small sidewall mass movement. The abrupt end of elevated SSC supports the hypothesis of an introduced sediment slug passing down the gully channel. The chronological Q-SSC plot (Figure 5.15(c)) exhibits a narrow, clockwise hysteresis loop, representing sediment supply depletion during SSC Peak 1, followed by clockwise loop, representing the SSC Peak 2 falling-limb sediment pulses. The arithmetic SSC scale in the Q-SSC plot better illustrates the sharp increase in SSC at 270 l/s rising-limb discharge than does the logarithmic scale in Figure 5.15a. 130 b) Temporal Frequency Distribution of SSC Exceedence 1000 c) Chronological Q versus SSC 100 80 |> 60 co 40 CO n Si : ": " O cs O SSCPeak2 100 200 Q (l/s) , SSC Peak 1 300 400 Figure 5.15. Discharge (Q) and suspended sediment concentration (SSC) at Gauging Station B during Runoff Event 2, 2 - 3 November, 1993. a) Q and SSC versus time, b) Temporal frequency distribution of SSC exceedence during runoff event, c) Chronological Q versus SSC plot. 131 5.5.3 RunoffEvent 6 at Gauging Stations B and C: The 2 - 3 December runoff event was generated by 108 mm of rainfall, probably accompanied by variable accumulation and melting of a shallow snowpack. Only the minor RunoffEvent 5 had occurred during the interval following the late November cP-anticyclonic cold spell and preceding RunoffEvent 6. At both gully gauging stations, event hydrographs consisted of a steep rising limb, a moderately steep falling limb, and a main hydrograph peak bounded on either side by a "shoulder" of steady, slightly-lower-than-peak discharge (Figures 5.16 and 5.17). At Gauging Station B, discharge during both shoulders and the main peak was in excess of the 270 l/s "replenished supply" threshold for about 18 hours. At both stations, three suspended sediment transport pulses — labelled SSC Peaks 1, 2 and 3 — occurred in association with the hydrograph rising-limb shoulder, peak, and falling-limb shoulder, respectively. Relative magnitude of the three SSC peaks differed between the two stations. Gauging Station B: At Gauging Station B, the highest SSC occurred during SSC Peak 1. SSC started to increase early on the hydrograph rising limb, reaching 40 mg/l by the onset of threshold discharge, the same SSC value as occurred at rising-limb threshold-discharge in RunoffEvent 2. From this, we can infer that sediment supply conditions were similar at the onset of Events 2 and 6 in Gully B. In contrast, Runoff Event 4 peak discharge was just barely sub-threshold (261 l/s), yet maximum SSC was less than 2 mg/l. Therefore, frost action during the cP-anticyclonic cold spell following RunoffEvent 4 probably replenished sediment supply, and the replenished supply was not entirely depleted by the flows of Runoff Event 5. During SSC Peak 2, which occurred in conjunction with the hydrograph peak, SSC reached about 70% of SSC Peak 1 value. SSC Peak 3, during the falling-limb hydrograph shoulder, was minor (15% of SSC Peak 1). 132 a) Discharge (Q) and Suspended Sediment Concentration (SSC) versus Time 1000 100 10 Q = 270 l/s Rising-Limb Shoulder: Hydrograph I eak Falling-Limb Shoulder 18:00 02 Dec. 00:00 03 Dec 06:00 12:00 Time (PST) 100 1000 O co co < M 00:00 04 Dec. • SSC b) Temporal Frequency Distribution of SSC Exceedence SSC (mg/l) c) Chronological Q versus SSC SSC Peak 1 SSCPeak2 200 300 . Q (l/s) 400 500 Figure 5.16. Discharge (Q) and suspended sediment concentration (SSC) at Gauging Station B during RunoffEvent 6, 2 - 3 December, 1993. a) Q and SSC versus time, b) Temporal frequency distribution of SSC exceedence during runoff event, c) Chronological Q versus SSC plot. 133 a) Discharge (Q) and Suspended Sediment Concentration (SSC) versus Time Hydrograph Peak 06:00 12:00 18:00 00:00 Time (PST) b) Temporal Frequency Distribution _ of SSC Exceedence . j?6 i SSC (mg/l) c) Chronological Q versus SSC 60 50 e-»o D) £ 3 0 o CO w 20 10 0 f SSC Peak 2 25 50 75 100 125 Q (l/s) Figure 5.17. Discharge (Q) and suspended sediment concentration (SSC) at Gauging Station C during RunoffEvent 6,2-3 December, 1993. a) Q and SSC versus time, b) Temporal frequency distribution of SSC exceedence during runoff event, c) Chronological Q versus SSC plot. 134 The chronological Q-SSC plot (Figure 5.16c) exhibits a wide, clockwise hysteresis loop, representing SSC Peak 1 depletion of sediment supply, especially the lower channel-margin sediment storage that was presumably replenished by frost action. The arithmetic SSC scale in the Q-SSC plot better illustrates the increase in SSC that occurred at 270 l/s rising-limb discharge during SSC Peak 1 than does the logarithmic scale in Figure 5.15a. A single-valued spike follows the clockwise loop, representing unlimited sediment supply during the peak flows associated with SSC Peak 2. Lower absolute SSC during the hydrograph peak was due to the prior removal of lower channel-margin material during SSC Peak 1. Gauging Station C: At Gauging Station C, SSC Peaks 1 and 3, occurring during the rising-limb and falling-limb hydrograph shoulders, respectively, were minor: less than 10 mg/1. SSC Peak 2 was the main transport pulse and was associated with the hydrograph peak! The chronological Q-SSC plot (Figure 5.17c) exhibits a moderately wide, clockwise hysteresis loop during SSC Peak 2, indicating depletion of sediment available at high flows. However, the depletion of frost-replenished sediment supply during low rising-limb flows which was observed at Station B was absent at Station C. 5.5.4 Runoff Event 7 at Gauging Station B: The 9 -10 December runoff event was generated by 102 mm of rainfall and melting of a shallow snowpack. At Gauging Station B, the hydrograph was single-peaked, with a steep rising limb and gentler falling limb (Figure 5.18). Maximum discharge exceeded 270 l/s for about 6 hours. One main SSC peak occurred in conjunction with the hydrograph peak. 135 b) Temporal Frequency Distribution _ of SSC Exceedence SSC (mg/1) c) Chronological Q versus SSC c > = 270 l/s 50 100 150 200 250 300 Q (l/s) Figure 5.18. Discharge (Q) and suspended sediment concentration (SSC) at Gauging Station B during Runoff Event 7, 9 - 10 December, 1993. a) Q and SSC versus time, b) Temporal frequency distribution of SSC exceedence during runoff event, c) Chronological Q versus SSC plot. 136 SSC at the onset of threshold discharge on the rising limb was only about 10 mg/l, as compared to 40 mg/l at the corresponding flow during Event 6, evidence of depleted sediment supply. The chronological Q-SSC relation exhibits a moderately broad, clockwise hysteresis loop, indicating further depletion of an already depleted sediment supply. Once again, the rising-limb portion of the Q-SSC plot is kinked upward at 270 l/s discharge, but the overall slope of the Q-SSC plot is much flatter than the Event 6 Q-SSC plot. 5.5.5 Runo f fEven t 11 at Gaug ing Stations B and C : The runoff event of 27 February - 3 March was generated by 149 mm of rainfall on a deep snowpack. This was the first major runoff event following the cP-anticyclonic cold spell of late January - early February. At both gully gauging stations, the hydrographs were composed of three discharge peaks, each successively higher than the last (Figures 5.19 and 5.20). At Gauging Station B, all three peak discharge values exceeded the 270 l/s replenished-supply discharge threshold. At both gauging stations, an SSC peak was associated with each hydrograph peak and was labelled accordingly. Event-maximum SSC at both stations was associated with the event-maximum discharge of Hydrograph Peak 3, and a period of sustained SSC occured on the falling limb of the event hydrograph. At Gauging Station B, additional SSC peaks occurred early on the event rising limb (SSC Peak la), on the falling limb of Hydrograph Peak 1 (lb), and on the rising limb of Hydrograph Peak 2 (2a); and sustained periods of elevated SSC occurred on the falling limb of Hydrograph Peak 2 and the rising limb of Hydrograph Peak 3. Gauging Station B: The SSC pattern at Station B appears to consist of four component patterns. First, the SSC pulse early on the event rising limb — SSC Peak la — was probably due to frost action 137 XI c CD 3 Si Go' <-»• 3. cr c i-f o* 3 O oo C/3 O CD X a a n Cu CD 3 o CD Cu C 3. 3 OQ a 3 O 0} CD < cD 3 o o 3 O O «B. o p O >i CO C CO 00 n o <—»• CO Cg ft O co O D* P o3 <D O a. CO c CO CD 3 Cu CD Cu CO CD Cu 3* CD 3 o o 3 O CD 3 s o 3 00 o Duration of SSC Exceedence (hr.) o cn o ui S K 8 o -I 1 1 (-f 3 OQ 00 f» f-t-o* 3 n Cu c a. 3 OQ 3 O m < CD 3 - J 8-S 5 o 3* VO VO o 3 _<?«_ 00 oo o < CD "I co C co CD H CD 3 X} O cr H re 3 o •a =»> o C/3 n p C/l n T I M 3 M ja re e rt re a. 3 o en re 2 CD 35* 3. cr B s* SSC (mg/l) © © o p M OC *) re cr o o M -I SSC (mg/l) 8 8 Rainfall at Russ2 (mm/hr) Q(l/s) Ui CO o D to CO o O 8 sr •8 s c c •a re s c. C/3 re a. §" re 3 n o a re re 3 C/3 C/3 re -i 3 re SSC (mg/l) u> VO replenishment of near-channel sediment supply prior to the onset of Runoff Event 11. Second, SSC of about 15 to 20 mg/1 was sutained whenever discharge remained above threshold; the 270 l/s threshold has been plotted in Figure 5.19, but the rapid SSC decline at the end of SSC Peak 1 and at the start and end of Peaks 2 and 3 coincides with about 290 l/s discharge. Third, random sediment transfer events were probably the cause of the anomalous SSC Peaks lb and 2a. Fourth, the highest SSC value occurred during the highest discharge, due to accessing of sediment supply that had not been subject to flow since RunoffEvent 6. The Q-SSC plots of the main hydrograph peaks all show narrow clockwise hysteresis, indicating minor sediment-supply depletion during event. Ganging Station C: At Gauging Station C, as at Station B, SSC maxima were associated with the three hydrograph peaks, and once again Peak 3 resulted in the event-maximum SSC value (Figure 5;20). In contrast to Station B, however, only a minor SSC rise was associated with onset of Peak 1 rising limb, and no SSC pulse occurred at the start of Peak 2. This indicates that pre-event recharge of channel-margin sediment supply was less in Gully C than in Gully B. Another difference: sustained periods of elevated SSC did not occur at Station C until after Hydrograph Peak 3, when some new sediment transfer apparently began to occur — such as sliding of melting snow covered with organic material into the channel. The individual-peak chronological Q-SSC plots show only minor supply depletion, as at Station B. The random SSC peaks preceding and between hydrograph peaks which can be seen in the Station B Q-SSC plot are not seen in the Station C plot. 140 5.5.6 RunoffEvent 13 at Gauging Stations B and C : The runoff event of 27 March - 1 April was generated by radiation snowmelt. At both gully gauging stations, Hydrograph Peaks 1 through 5 occurred daily, in late afternoon to early evening, in response to daytime snowmelt, although the diurnal cycles were highly muted at Station C compared to Station B (Figures 5.21 and 5.22). The greater diurnal discharge fluctuation at Station B was probably due to lack of forest cover in Gully B. Gully B would have received greater solar insolation by day and radiatively cooled more efficiently by night, resulting in greater diurnal fluctuation in snowmelt rates. At both stations, event-maximum discharge occurred on the evening of 28 March (Hydrograph Peak 2), the day with the highest maximum temperature (25°C). At Gauging Station B, SSC maxima were associated with each hydrograph peak. The SSC maximum values increased on consecutive days, from Peak 1 through Peak 3, despite the peak discharge of Peak 3 being less than Peak 2. Chronological Q-SSC pattern within Hydrograph Peaks 1 and 2 was clockwise hysteretic, but within Peak 3 the pattern was counter-clockwise hysteretic. The pattern during Peaks 1 and 2 represents the usual multiple-peak depletion-within-peaks pattern. However, sediment supply replenishment must have occurred between Peaks 2 and 3, probably associated with snowmelt. The large snowmelt of 28 March (Peak 2) would have left fresh sediment available for mobilisation on the rising limb of Peak 3. In addition, more sediment was evidently released later on 29 March, contributing to the counter-clockwise hysteresis loop observed in the Peak 3 Q-SSC plot. Snowmelt could cause such a cycle of sediment-supply replenishment and depletion if sediment were held within the snowpack at the base of hillslope sediment sources or if sediment sources were protected from erosion by an overlying snow cover. In either case, snowmelt would allow a new sediment supply to be mobilised by snowmelt runoff. A threshold must have been reached late on 28 March, when a 141 a) Discharge (Q) and Suspended Sediment Concentration (SSC) versus Time 1000 100 4 o co co 12:00 Time (PST) b) Temporal Frequency Distribution of SSC Exceedence £ 5 SSC (mg/l) Figure 5.21. Discharge (Q) and suspended sediment concentration (SSC) at Gauging Station B during RunoffEvent 13, 27 March - 1 April, 1994. a) Q and SSC versus time, b) Temporal frequency distribution of SSC exceedence during runoff event, c) Chronological Q versus SSC plot. 142 b) Temporal Frequency Distribution _ of SSC Exceedence £ 125 S S C (mg/l) Figure 5.22. Discharge (Q) and suspended sediment concentration (SSC) at Gauging Station C during RunoffEvent 13, 27 March - 1 April, 1994. a) Q and SSC versus time, b) Temporal frequency distribution of SSC exceedence during runoff event, c) Chronological Q versus SSC plot. 143 relatively large supply of sediment was exposed. Exposure then continued through 29 March. Depletion of the newly exposed sediment supply probably did not occur during Peak 3. Declining SSC maxima on 30 March and onward were probably the result of restricted access to the new sediment supply due to falling daytime temperatures and consequent lower snowmelt rates and smaller daily peak discharges. At Gauging Station C (Figure 5.22), the temporal SSC pattern differed greatly from Station B. SSC was high on Hydrograph Peak 1 rising limb, and maximum SSC of the runoff event (52 mg/1) occurred almost in phase with maximum discharge of Peak 1. After the Peak 1 SSC maximum, SSC dropped to an almost constant value of 20 to 30 mg/1 for the next three days, finally dropping below 10 mg/1 on 1 April. The long-lasting, moderately-high SSC episode of 28 -31 March was unique within the study period at either station. This episode resulted in the largest single runoff-event suspended sediment transport amount of the study period at Gauging Station C. Some new sediment source must have been exposed by snowmelt, or else some sediment transfer mechanism associated with snowmelt runoff entering the gully-channel must have occurred. Perhaps organic material within the channel-margin snowpack was nearly continuously released during snowmelt.- Whatever the source, the sustained uniformity of elevated SSC during Event 13 remains unusual. 5.5.7 RunoffEvent 15 at Gauging Station B: The 10-12 April runoff event resulted from rainfall, probably falling on and melting a shallow snowpack. At Gauging Station B, the event hydrograph was standard single-peaked (Figure 5.23). Suspended sediment transport was relatively low until discharge reached the 270 l/s threshold which produced a short burst of high SSC. High-SSC burst duration was much briefer than the 4- hour duration of threshold-exceeding flow. Transport dropped rapidly 144 b) Temporal Frequency Distribution of SSC Exceedence £ 5 100 SSC (mg/l) 1000 Figure 5.23. Discharge (Q) and suspended sediment concentration (SSC) at Gauging Station B during RunoffEvent 15, 10 - 12 Apri l , 1994. a) Q and SSC versus time, b) Temporal frequency distribution of SSC exceedence during runoff event, c) Chronological Q versus SSC plot. 145 afterwards. Apparently, sediment supply had largely been depleted along the gully channel prior to RunoffEvent 13, but sediment storage remained at higher elevations above the channel, just barely accessed by the peak flows of Event 13. The chronological Q-SSC plot highlights the existence of the 270 l/s discharge threshold. The flat Q-SSC relation below 270 l/s discharge indicates that sub-threshold sediment supply was already depleted before event onset. Q-SSC pattern is clockwise hysteretic above 270 l/s, with SSC dropping to near sub-threshold level before hydrograph peak, indicating depletion of upper-elevation sediment supply occurred during the event. 5.5.8 S u m m a r y Suspended sediment transport within runoff events at Gauging Station B generally occurred dominantly on hydrograph rising limbs, resulting in clockwise Q-SSC hystereses, due to depletion of channel-margin sediment supplies. When discharge exceeded 270 l/s, however, sediment supply was essentially unlimited. Minor sediment supply replenishment — probably due to frost action — occurred after cP anticyclonic cold spells. More significant replenishment due to snowmelt processes occurred throughout the snowmelt period in Sub-season III, even between successive hydrograph peaks within runoff events. At Gauging Station C, Q-SSC plots were usually clockwise hysteretic in sub-seasons I and II, but the magnitude of early rising-limb sediment flushing was less than at Station B. No unlimited sediment supply discharge-threshold was observed. Sediment-supply recharge by frost action appeared to be less significant at Station C than at Station B, while snowmelt processes seemed to be more significant at Station C than at Station B. At both gully gauging stations, suspended sediment transport occurred during relatively brief, discrete episodes. 146 5.6. Debris-torrent transport in Gully B A debris torrent occurred in Gully B in the mid-afternoon of 25 October, 1994. Post-torrent surveys were utilised to make a rudimentary comparison of relative fine-sediment yields in Gully B by debris torrent versus suspended sediment transport. Gauging Station B instumentation was still operating prior to the debris torrent but was partially buried and rendered useless during the event. Fortunately, the instruments were salvaged. Local MacMillan-Bloedel personnel reported that heavy rain had fallen on a fresh snowpack in the 24 hours prior to the debris torrent. Weather data from station Russ2 are not available during this period. At Port Hardy Airport, 92 mm of rain fell on 25 October. By comparison, the largest daily rainfall at Port Hardy Airport during the study period was 76 mm. Estimated return period of the 25 October daily rainfall total is two to five years (Environment Canada rainfall intensity-duration frequency curves). Russell Creek discharge peaked at 22.7 m3/s early in the morning of 25 October. By comparison, Russell Creek peak discharges during the two largest study-period runoff events were 23.2 and 22.5 m3/s. Therefore, the 25 October, 1994, runoff event can be described as moderately large, but not extremely large. Turbidity and SSC data are not available from Russell Creek during this runoff event. The debris torrent was apparently triggered by a gully-sidewall debris slide. The slide occurred in the previously dormant sidewall depression between failure scars "F5" and "F6" (see Figure 2.6 for location). The debris torrent transported the debris-slide material down Gully B, entraining material stored in the gully channel along the way. The gully channel had been scoured to bedrock along most of its length above the big LOD jam by previous debris torrents. Some 147 material was removed from the big sediment wedge and a channel was scoured around the big LOD jam, but the jam structure was not destroyed. The gully channel below the big LOD jam, which previously stored considerable sediment, was scoured along its length to the depositional zone below road TS120. A new LOD jam, about 2 m high, formed where the torrent entered the forest, just upstream of Gauging Station B (Figure 5.24). The main depositional sediment wedge occurred behind the new jam. Volume of the initiating sidewall slide was estimated by post-slide survey to be 250 m3. The estimated sediment volume removed from the big pre-torrent LOD wedge was 30 m3. Gully-channel scour below the big pre-torrent LOD jam eroded approximately 350 m length by 5 m width by 1 m depth, for a total volume of 1750 m3. Thus the total sediment volume of the debris torrent was approximately 2030 m3, eight times the volume of the initiating slide. Assuming a sediment bulk density of 1500 kg/m3, total sediment mass transported to the depositional zone was about 3 x 106 kg. However, the suspendible sediment fraction of debris-torrent transport (fine sand and finer) probably consisted of about 10% of the sidewall slide mass -- 3.8 x 104 kg ~ plus 20% of the sediment-wedge scour material mass -- 0.9 x 104 kg — for a total of 4.7 x 104 kg. I have assumed that sediment stored within the channel, other than behind the big L O D jam, was free of suspendible fines, these having been removed by normal-regime fluvial transport (as discussed in Section 1.3.3). Fines percentages in sidewall-slide and sediment-wedge material are based on particle-size distribution data provided in Appendix C-2. By comparison, suspended sediment transport at Gauging Station B during the entire 1993-94 data-collection period amounted to 2.9 x 103 kg — about 6% of the fine-sediment mass transported during the debris torrent. Thus, the debris torrent transported the equvalent fine-sediment load as approximately 16 years of normal-regime fluvial transport in Gully B, assuming 148 Figure 5.24. Gully B debris-torrent deposit, November 1994. a) Looking downstream at fresh LOD jam which formed at the forest edge. Gauging Station B located about 20 m beyond jam. b) Looking upstream from same point. 149 that truly "normal" conditions were observed during the 1993-94 study period and that the October 1994, debris torrent was typical in magnitude. Concurrent research being conducted by Shannon Sterling on debris torrent characteristics in Tsitika Watershed indicates that in torrent-prone gullies, inter-torrent intervals typically range from about 5 to 30 years (Sterling, pers. comm.). Gully B was not included in the Tsitika debris torrent study. However, field observations indicate that at least one other debris torrent had occurred in Gully B since the gully was logged in the early 1980's. Dormant sidewall depressions with stumps of mature trees on them indicate that sidewall-failure recurrence interval of individual depressions may be on the order of many decades. Based on these observations, Gully B debris-torrent recurrence interval is probably in the range of several years to several decades, falling within or near the typical Tsitika range. Therefore, cursory analysis suggests that long-term fine-sediment yield by fluvial and debris-torrent transport in Gully B may be of similar magnitude, or at least that the two values probably lie within an order of magnitude of one another. 150 Chapter 6 Suspended Sediment Transport in Russell Creek Suspended sediment transport results from Russell Creek Gauging Station are presented in this chapter. Table 6.1 summarises the portion of the study period during which data were successfully collected. Study-period discharge and SSC are displayed in Figure 6.1. More detailed 20-day plots of Russell Creek discharge and SSC may be found in Appendix A-4. Eighteen independent runoff events were identified at Russell Creek Gauging Station during the study period, 17 of which coincide with runoff events identified at the gully gauging stations. The hydrologic and sedimentologic characteristics of the runoff events are summarised in Table 6.2. Table 6.1. Summary of data collection at Russell Creek Gauging Station Period Duration (days) Fraction of study period (%) Study period: 30 Sep. /93 - 27 Jun. /94 270 100% Discharge data collected 270 100% Discharge and SSC data collected 132 49% Notes: 1. "Discharge data collected" includes daily estimates provided by WSC. 2. SSC data collection truncated in the latter stages of the two largest runoff events, Events 6 and 11. Suspended sediment transport values calculated for these events are underestimates. 151 Q (ma's) S S C (mg/l) to Table 6.2. Summary of runoff events at Russell Creek Gauging Station Event Date Runoff Suspended Sediment Transport # (1993-94) Max. (1) Total Runoff Max. 15-min. Total Transported Discharge (m3/s) Volume (x 10 A 6 m3) Cone, (mg/l) Mass (x 1000 kg) 1 21 - 24 Oct. 4.7 0.65 9 0.2 2 01 - 04 Nov. 11.4 1.44 M M 3 1 4 - 1 7 Nov. 10.2 1.70 9 4.1 4 1 9 - 2 2 Nov. 8.0 1.41 7 2.7 5 2 8 - 3 0 Nov. 4.4 0.62 9 0.7 6 01 - 04 Dec. 23.2 2.99 536 95.7 7 0 8 - 1 5 Dec. 13.4 4.53 M M 8 28 Dec. - 05 Jan. 9.5 1.97 220 2.0 9 1 2 - 15 Jan . 3.8 0.67 20 0.2 10 21 - 24 Jan . 6.4 0.97 0 0.2 10a 1 2 - 1 8 Feb. 4.3 0.95 15 0.5 11 27 Feb. - 04 Mar. 22.5 4.56 483 96.5 12 1 2 - 1 4 Mar. 4.2 0.62 M M 13 26 - 30 Mar. 5.2 1.08 M M 14 0 1 - 0 3 Apr. 3.7 0.47 M M 15 11 - 1 2 Apr. 8.9 0.73 M M 16 1 4 - 2 0 Apr. 4.6 1.11 M M 17 1 2 - 1 5 Jun. 3.3 0.57 M M cn-633.wb1 Notes: . 1. Discharge values are estimated daily averages for Events 1 through 7, and recorded 6-hour averages for Events 8 through 17. 2. " M " = missing data 153 6.1 Hydrological comparison of Russell Creek Gauging Station to gully gauging stations Seventeen of the runoff events identified at Russell Creek Gauging Station coincide with the 17 runoff events identified at the gully gauging stations. In addition, one more runoff event ~ Iabelled "10a" -- occurred at the Russell Creek Station in mid-February as a result of a storm in which the rain/snow transition was at about 500 to 600 m elevation (approximate lower elevations of Gullies B and C). Peak discharge and total runoff volume during study-period runoff events at Russell Creek Gauging Station and the gully gauging stations are compared in Figure 6.2. Moderately strong, positive relations exist between hydrologic characteristics of runoff events in Russell Creek and the study gullies, especially among larger runoff events. In Figure 6.3, areally-normalised plots of runoff-event hydrologic characteristics show that specific peak discharge and runoff volume are considerably less in Russell Creek than in the study gullies, as expected. 6.2 Importance of runoff events in transporting suspended sediment Table 6.3 summarises total recorded runoff volume and suspended sediment transport during the study period and during study-period runoff events at Russell Creek Gauging Station. Temporal distribution of flow in Russell Creek was similar to flow distribution in gullies. Runoff-event duration occupied about one-quarter of the study period, and runoff contribution was over 80% of study-period total. Suspended sediment transport during runoff events accounted for 99% of total study-period transport, an even greater fraction than in the study gullies. Since data collection periods at the three gauging stations are not identical, comparisons of total runoff and sediment transport are not truly valid but they do indicate pattern similarities. In any case, suspended sediment transport at Russell Creek Gauging Station occurred primarily during runoff events; therefore, analysis will focus on runoff events only. 154 a) 1000 5. 100 4-S 104-CL Peak Discharge -••i i 13 -7... 4 M S i. 2 9 " - - V urn ' » • • . 3 1 7 " 12 16 . . 13 < . . •* . 11 10 14 1+r + - * 3 + 1 4 ::;::12 ::::i*:!:::i::j::|:::::: : . i io. 2 : • .16 < i • .4 : j 5 ;....'....;...].. - i — 1 7 ±—i—1—i—i—i-10 Peak Q, Russell (m3/s) Stn. B +• Stn. C 100 b) 100 CO E o § ,0 x 3 (5 o c a: 0.1 0.1 Runoff Vo lume - ' 1 10 ' 13 T""'9 4 • 8 • 11: 11 12 H-15 .+ 13 + 6 A4^L...:±L.\..;^.., .....±3 ::j::::::::::::j::::::::i...::?::::!l ::::; ••] • 1 \ >--M j-4--i i i j-44-n-M-: 10 + •16 H 1 1 1—I I i Runoff Vol., Russell (x 10A6 m3) 10 Stn. B + Stn. C Figure 6.2. Hydrological comparison of Russell Creek Gauging Station to Gauging Stations B and C. Comparison of a) peak discharges, and b) runoff volumes, during the study-period runoff events for which flow data are available. 155 0.1 H 1 1 1 1-Spec. Peak Q, Russell Cr. (l/s/ha) 10 Stn. B •+- Stn. C b) 10 to «£ CO E 3 8. 0.1 0.1 Specific Event Runoff Volume - i i Spec. Vol., Russell (x 1,000 m3/ha) Stn. B + Stn. C 10 Figure 6.3. Areally-normalised hydrological comparison of Russell Creek Gauging Station to Gauging Stations B and C. Comparison of a) specific peak discharges, and b) specific runoff volumes, during the study-period runoff events for which flow data are available. 156 Table 6.3. Importance of runoff events in transporting suspended sediment at Russell Creek Gauging Station Discharge and SSC data collection period: Total runoff volume (x 10A6 m3) 20.4 Total suspended sediment transport (kg) 204.1 Runoff events with recorded discharge and SSC data: Number of runoff events with recorded discharge and SSC data: 10 Duration of runoff events (days): 32 Duration of runoff events as fraction of discharge and SSC data collection period (%): 24% Total runoff volume (x 1,000 m3): 16.5 Runoff volume during runoff events as fraction of data collection period total: 81% Total suspended sediment transport (kg): 202.6 Suspended sediment transport during runoff events as fraction of data collection period total: 99% cn-633.wb1 6.3 Seasonal scale patterns SSC data collection period at Russell Creek Gauging Station covers the late 1 9 9 3 dry season and the first two sub-seasons of the 1 9 9 3 - 9 4 wet season. SSC data from most of Sub-season III and the early 1 9 9 4 dry season are missing. Event-by-event display of peak discharge, peak SSC, runoff volume, and suspended sediment transport at Russell Creek Gauging Station is provided in Figure 6 . 4 . As at the gully gauging stations, no seasonal decline in suspended sediment transport is evident in Russell Creek, during the portion of the wet season in which SSC data were collected. Runoff-event suspended sediment transport is plotted against peak discharge in Figure 6 . 5 to investigate the possibility of a peak-discharge threshold, above which transport increases greatly. Runoff Event 6 and 11 transport values plot below the best-fit discharge-transport relation, but SSC data collection was truncated in the latter stages of both events; therefore, true transport values are greater than those shown in Figure 6 . 5 . Suspended sediment transport is closely related to peak discharge at Russell Creek Gauging Station. No peak-discharge threshold is apparent. Figure 6 . 6 shows suspended sediment transport versus runoff volume for all events. A moderately strong relation exists, expressed as: T = 0 . 8 V 3 " where T = suspended sediment transport (x 1 0 3 kg), V = runoff volume (x 1 0 6 m3), for individual independent runoff events. For comparison, the gully gauging station transport-runoff relations are included in Figure 6 . 6 . Two main differences exist between the gully relations and the Russell Creek relation. First, 158 a) Runoff Volume and Susp. Sed. Transport — 4 co E co < X, o E .2 •5 2 o c a 1 CO ..to .. 1 -c r CO 7 -t J... 1 2 3 4 5 6 7 8 9 1010a11 12 13 14 15 16 17 Runoff Event • Runoff Vol. j Susp. Sed. Transp. -100 -80 -60 - < — M (SSQ-[*--420 c i 4 4 0 £ "8 CO b) CO E. <D 30 25 20 15 10 5 0 Peak Discharge and S S C -o 4 - , CO CO - M (SSC) 1 2 3 4 5 6 7 8 9 10 10a11 12 13 14 15 1 Runoff Event 6 17 | | Discharge ^ SSC 4 4 0 0 e cn E 600 500 300 200 O co CO ra a> 0. 4 1 0 0 Figure 6.4. Event-by-event display of flow and suspended sediment transport characteristics during study-period runoff events at Russell Creek Gauging Station, a) Runoff volume and suspended sediment transport, b) Peak discharge and suspended sediment concentration. 159 Russe l l Creek Gaug ing Stat ion Peak Discharge (m3/s) Figure 6.5. Comparison of suspended sediment transport to peak discharge during study-period runoff events at Russell Creek Gauging Station. the combined data from the two gully stations fall into two bands, which were explained in terms of sediment supply regimes in Chapter 5. On the other hand, sediment supply patterns are not evident at Russell Creek Station during the data collection period. At the gully gauging stations, subtle patterns were found within Sub-seasons I and II, followed by a more obvious difference between the first two sub-seasons and Sub-season III. Unfortunately, the latter result cannot be compared to Russell Creek due to lack of data. Another difference: the slope of the median transport relation at Russell Creek Station is much steeper than at the gully stations. Slopes of the "replenished supply" and "depleted supply" transport-runoff relations for the gully stations were 0.95 and 0.69, respectively. These convert to -0.05 (approximately zero) and -0.31 for mean concentration-runoff relations. In other words, 160 b) Gauging Stat ions B and C 10000 Runoff Vol. (x1000 m3) 100 Stn. B + Stn. C Figure 6 .6 . Comparison of runoff volume and suspended sediment transport during study-period runoff events at a) Russell Creek Gauging Station, and b) Gauging Stations B and C. 161 during periods in which sediment supplies are replenished, event-mean SSC is approximately constant regardless of runoff volume; during depleted supply periods, mean SSC decreases with increasing runoff volume, essentially due to dilution of available sediment. In Russell Creek, however, slope of the transport-runoff relation is 3.57, or 2.57 for mean SSC-runoff relation. The steep slope means that large runoff events in Russell Creek transport disproportionately large amounts of suspended sediment. Suspended sediment transport is probably more steadily related to runoff in Russell Creek than in the study gullies because in the larger Russell Creek system alluvial sediment storage in-channel and adjacent-to-channel is a more important contributor to sediment load than are hillslope and gully-channel inputs. Mobilisation of alluvium is more closely related to discharge than are transfers of hillslope sediment to stream channels (either directly or via gully channels). Transport probably increases more rapidly with increasing runoff in Russell Creek than in the gullies because alluvium in Russell Creek represents an almost unlimited sediment supply which requires flow energy to be mobilised. In the gullies, sediment supplies are more spatially limited, even in Gully B where the gully channel is lined with storage aprons and cones. Despite the abundance of these, flow access to them is highly restricted. As the storage units creep into the channel, they are quickly trimmed off. 6.4 Synoptic scale patterns Characteristics of suspended sediment transport within the 10 runoff events for which Russell Creek data are available are presented in Table 6.4. Three events were selected for synoptic-scale study. 162 r o Q . VI B cs u T«l s s •3 cu Vi •a CU T3 B <U CU Vi 3 vi B C/3 O X ) _c "5b s cs O CU CU u CU u cs i_ es JS cj « u i _ C J O. o B C/3 VO <u 2 es H s 03 to B CU > <u ta o B S 0X1 s • mm u 3 •o Chronological Q vs. SSC relation type (3) 1 I 1 1 CM CM • co 1 2 1 1 M-2/3 1 insport ;nce duration it duration 20 mg/l o _ o o o o o O CO —^ to o o O CO o* o o § 3 2 2 2 2 2 . 2 Sediment Trc SSC exceede as % ever 100 mg/l | o _ o o o d o o o> _ Tt o o o Tt o o o 2 ^ 2 2 2 2 2 2 Suspended of SSC nee (hr.) 20 mg/l o o o ° 5 ° . ° o o o o o o o o o ui 5 o q o O Tt ^ O o o o <=>. °. 2 2 2 2 2 o m 2 Duration exceede 100 mg/l | 0.00 M 0.00 0.00 o o m o o O CO o o o o o °. °. 2 2 2 2 2 O Tt 2 Max. 15-min. SSC (mg/l) o i 5 oi s CD O Q oi co 5 eg y o lO CM £ S 2 2 2 2 2 •*TT 2 Duration (1) (hr.) CO «5 CM CN CO CN ^ 2 O CM M- N ^ ® CD S 5 CM 2 2 2 2 2 2 :  Event Generation Type i - i - CM CM CM CM CM CM CO CM CM CM CM CO CM CM CO Runofl Date (1993-94) 22 - 24 Oct. 01 - 04 Nov. 14-17 Nov. 19-22 Nov. 29- 30 Nov. 02 - 04 Dec. 09-11 Dec. 29 Dec. - 05 Jan. 12-14 Jan. 01 . OA Ian £. I - £.*+ Jail. 12-18 Feb. 27 Feb. - 04 Mar. 12-14 Mar. 26 Mar. - 01 Apr. 01 - 03 Apr. l o - i £. Apr. 14 -19 Apr. 12-13 Jun. c LU •<- CM co -<t i o co s co cn ° § T- CM co Tt m co X T3 C CU a . C L < cn" 3 co Q . ^ CD CU E? 5 co co >. cu 5 H O t/> T- CU E £ o T3 cu c cu •o cn c o _C0 cu k_ O C O C O c/i > ~ co O T3 CO cu o z 3 CD CO of 163 6.4.1 RunoffEvent 6 at Russell Creek Gauging Station Hourly discharge data were not available from Russell Creek Gauging Station during the 2-3 December runoff event; however, estimated daily discharges were provided by WSC (Figure 6.7). At the daily scale, the hydrograph was single-peaked. Daily discharge on 3 December was the highest of the study period. Suspended sediment transport began suddenly on the morning of 3 December, probably on the hydrograph rising limb (comparing timing to gully stations). SSC data were truncated partway through the event while SSC values were still above 100 mg/1. Due to data limitations, a Q-SSC pattern cannot be defined for RunoffEvent 6. However, the SSC rise was rapid, similar to that noted at the gully stations. SSC samples were obtained from Russell Creek by pumping sampler during Runoff Event 6. In Figure 6.8, sampled SSC values are compared to the continuous SSC estimate based on lab-calibration of the turbidity-SSC relation. Two points are evident from the comparison. First, the similarity of SSC magnitudes obtained by the two methods — if average trend and magnitude range are considered ~ supports validity of the turbidity-based lab-calibation formula. Second, variability in the turbidity-based SSC estimate is much greater than in the SSC samples. Although turbidity values were averaged over 15-minute intervals, and SSC samples were drawn over a period of only about one minute, the greater instability of turbidity was not damped by the longer sampling interval. The difference in turbidity and SSC variabilities is significant, so much so that a sample-based turbidity-SSC relation could not be used at Russell Creek Gauging Station. 6.4.2 RunoffEvent 8 at Russell Creek Gauging Station During the 29 December - 5 January multi-peaked runoff event, elevated SSC occurred only in a brief burst on the rising limb of the first hydrograph peak (Figure 6.9). Suspended 164 a) Discharge (Q) and Suspended Sediment Concentration (SSC) versus Time 12:00 15:00 18:00 21:00 00:00 03:00 06:00 . 09:00 12:00 02 Dec. 03 Dec. Date — — SSC Q (est. daily) b) Temporal Frequency Distribution of SSC Exceedence Figure 6.7. Discharge (Q) and suspended sediment concentration (SSC) at Russell Creek Gauging Station during RunoffEvent 6, 2 - 3 December, 1993. a) Q and SSC versus time, b) Temporal frequency distribution of SSC exceedence during runoff event. Chronological Q versus SSC not plotted since only daily discharge data available. 165 166 a) Discharge (Q) and Suspended Sediment Concentration (SSC) versus Time 100 10 E, o 0.1 29-Dec 30-Dec 31-Dec 01-Jan 02-Jan Date o CO co —I y-i—i—|——\— 03-Jan 04-Jan 05-Jan SSC Q (6-hr.) b) Temporal Frequency Distribution of SSC Exceedence V g 1.25 g 1.00 ' CO 2 0.50 o | 0.25 Q 0.00 1 ! ! ! II 1 i i i - -r 1 J 100 1000 SSC (mg/1) Figure 6.9. Discharge (Q) and suspended sediment concentration (SSC) at Russell Creek Gauging Station during RunoffEvent 8, 29 December, 1993 - 5 January, 1994. a) Q and SSC versus time, b) Temporal frequency distribution of SSC exceedence during runoff event, c) Chronological Q versus SSC plot. 167 sediment transported during this burst probably represented easily mobilised material that had been deposited among coarser bed-surface or bar-surface particles. . Bedload transport probably did not occur during Runoff Event 8 because SSC was negligible during the highest flows on 4 January when bedload transport would have been most likely to occur. A period of high SSC would probably accompany a bedload transport episode because the normal armour-layer protection of finer sub-surface material would disappear during armour-layer transport. The Q-SSC relation for Event 8 consists of a single, narrow, clockwise hysteresis loop, indicative of rapid sediment supply depletion. This is consistent with entrainment of a minor quantity of surface fines with no subsequent inputs from stream substrate or hillslopes and gullies. 6.4.3 RunoffEvent 11 at Russell Creek Gauging Station The hydrograph of the 27 February - 3 March runoff event is triple-peaked with maximum discharge increasing on each successive peak (Figure 6.10). SSC data collection was truncated near the time of Hydrograph Peak 3. Two main high-SSC episodes — labelled SSC Peaks 1 and 2 -- occurred during Runoff Event 11, in association with Hydrograph Peaks 1 and 3. During Hydrograph Peak 2, SSC barely rose above 10 mg/1. Q-SSC pattern during SSC Peak 1 was nearly single-valued, with slight counter-clockwise hysteresis. During SSC Peak 2, Q-SSC hysteresis was slightly clockwise. Suspended sediment transport during SSC Peak 1 may have been due to bed material entrainment or hillslope and gully inputs, or both. The slight lag of transport peak behind discharge peak during Hydrograph Peak 1 may be indicative of inertial effects in bed-material entrainment. Low SSC during Hydrograph Peak 2 may be explained by continuing bedload transport of "cleaned" gravels, all available fine sediment in the bed material having been flushed downstream during Hydrograph Peak 1. 168 a) Discharge (Q) and Suspended Sediment Concentration (SSC) versus Time b) Temporal Frequency Distribution of SSC Exceedence t ? 3 n , . §25 V T> S20 ^15 CO 2 10 0 1 5 nj 2 n I X | \ j L [ i i 1 i i • ! " 10 100 1 SSC (mg/l) 300 c) Chronological Q versus SSC 400 i ; ; -SSC Peak 2 -300 — 200 O co co 100 SSC Peak 1 ml 10 15 20 Discharge (m3/s) 25 30 Figure 6.10. Discharge (Q) and suspended sediment concentration (SSC) at Russell Creek Gauging Station during RunoffEvent 11, 27 February - 3 March, 1994. a) Q and SSC versus time, b) Temporal frequency distribution of SSC exceedence during runoff event, c) Chrono-logical Q versus SSC plot. 169 Suspended sediment transport during Hydrograph Peak 3 corresponds with the event-maximum period of suspended sediment transport at the gully gauging stations. Therefore, hillslopes and gullies may have provided this pulse of high SSC. Clockwise hysteresis also suggests input from gully flushing rather than from release of bed-material fines. 6.4.4 S u m m a r y At Russell Creek Gauging Station, suspended sediment transport within runoff events occurred in relatively brief, discrete episodes, similar to the bursts observed at the gully gauging stations. Based on comparison of only three runoff events, temporal transport patterns at Russell Creek Station were similar to those at the gully gauging stations, but a slight SSC lag may exist due to inertial effects in bed-material mobilisation. 170 Chapter 7 Discussion and Conclusions 7.1 Discussion of hypotheses Study hypotheses were listed in Section 1.6.2. In this section, the validity of each hypothesis is discussed with reference to results presented in Chapters 4 through 6. Hypotheses have been phrased generally, while results are study-specific. Hypothesis validity discussed below does not necessarily extend to the general case. 7.1.1 Hillslope to gully-channel transfers 1) At hillslope sediment sources which consist of non-cohesive banks or slopes, fine sediment is transferred to channels primarily during the dry season by dry ravel, during early wet-season rainstorms by rainsplash, and during winter cold snaps by frost action. Hypothesis 1 was partially sustained. Sediment transfer at monitored sediment-source Sites 2 (active sidewall), 5 (cross-ditch bank), and 6 (road cutbank) occurred most rapidly during periods in which needle ice development was observed and continued gradually during the dry season. These seasonal patterns are in agreement with the hypothesised sediment-transfer processes; however, relative importance of these processes, or the exclusive occurrence of these processes, cannot be proven from the rudimentary sediment-transfer data collected in this study. 171 2) At hillslope sediment sources which consist of thin sediment veneers over impermeble surfaces, fine sediment is transferred to channels primarily during wet-season rainfall and/or snowmelt events by sheetwash and rapid mass movements. Hypothesis 2 was partially sustained. Sediment transfer was monitored at two failure scars -- Sites 1 and 3. At Site 1, evidence suggests gradual accumulation of sediment on the failure scar was punctuated by shallow sediment-veneer slides, which occurred during at least two different episodes during the wet season. Site 3 was monitored for only three weeks, during which time considerable sediment was transported to the base of the scar, apparently by a transitional fluvial/colluvial mode. Once again, temporal patterns agree with hypothesised processes, but occurrence and relative importance of processes cannot be proven. 7.1.2 Suspended sediment transport in gullies : seasonal and sub-seasonal patterns 3) Suspended sediment transport in gullies occurs almost entirely during large rainstorm and/or snowmelt runoff events. Hypothesis 3 was sustained. At least 89% and 92% of suspended sediment transport at Gauging Stations B and C, respectively, occurred during runoff events. 4) Regardless of antecedent hydrometeorological conditions, the largest runoff events transport the greatest quantities of suspended sediment. Hypothesis 4 was sustained for Gauging Station B but not for Station C. At Gauging Station B, 65% of study-period suspended sediment transport occurred during the two runoff events with highest peak discharges, Events 6 and 11 . Furthermore, a sediment-supply discharge threshold of 270 l/s was identified by comparing transport between events and by analysis of 172 transport within events. The discharge threshold was barely exceeded by four events during the study period and greatly exceeded twice (Events 6 and 11). The 270 l/s threshold probably defines the flow which barely accesses unarmoured sediment-storage compartments along the gully channel-margin. Flows less than 270 l/s access only the coarsened lower flanks of channel-margin sediment-storage compartments. At Gauging Station C, in contrast, maximum suspended sediment transport — 37% of study-period total ~ occurred during RunoffEvent 13, which had the fifth-highest peak discharge of the study-period. Event 13 was characterised by a prolonged period of radiation snowmelt, moderately high discharge, and moderately high SSC. As at Station B, Runoff Events 6 and 11 had the highest peak discharges of the study-period at Station C and were next most important in terms of suspended sediment transport. Combined transport during Events 6 and 11 accounted for 47% of study-period total at Station C. No sediment-supply discharge threshold was identified at Station C, probably due to absence of channel-margin sediment-storage compartments. 5) By controlling replenishment of channel-margin sediment supply, antecedent hydrometeorological conditions influence suspended sediment output from gidlies during small to intermediate runoff events. Hypothesis 5 was sustained. At both gully gauging stations, suspended sediment transport during runoff events was clearly divided into two sediment supply regimes, referred to as "replenished supply" and "depleted supply". Snowmelt-related processes appear to be the dominant replenishment mode in both gullies. Frost action provided minor sediment supply replenishment in Gully B, but had negligible effect in Gully C. Transport during the first runoff event of the wet season at Gully C exhibited replenished supply. 173 Overall, antecedent conditions were more important in Gully C (stable sidewalls) than in Gully B (numerous sediment sources), in terms of controlling runoff-event suspended sediment transport, probably due to the buffering effect of channel-margin sediment storage compartments in Gully B. Transport in Gully B was most strongly controlled by peak discharge (ie. access to undepleted sediment supply). 7.1.3 Suspended sediment transport in gullies : synoptic scale patterns 6) Within runoff events not dominantly generated by snowmelt, Q-SSC relations exhibit clockwise hysteresis patterns due to depletion of channel-margin sediment supply during runoff events. Hypothesis 6 was sustained. Clockwise hystereses were the dominant Q-SSC pattern at both gauging stations prior to the main snowmelt period. Rising-limb sediment flushing was more pronounced at Station B than at Station C during Runoff Events 6 and 11. These were large events following periods of frost action in which the minor replenishment by frost action was obscured by the sheer magnitude of runoff and sediment transport, but the within-event analyses pointed out the early rising-limb differences. 7) At very high flows, single-valued Q-SSC relations may occur in gidlies with numerous hillslope sediment sources, due to accessing of unlimited channel-margin sediment supply. Hypothesis 7 was partially sustained. During Runoff Events 6 and 11. at Station B, Q-SSC relations were nearly single-valued at very high flows. This situation also occurred during Runoff Event 11 at Station C: Unexpectedly, Q-SSC was, single-valued during the relatively modest, steady discharge of Runoff Event 13 at Station C, marking this event as unique. In this case, 174 absence of sediment-supply limitation was probably not due to high flows accessing plentiful sediment adjacent to the channel, but rather to continued delivery of sediment to the channel by melting snow throughout the runoff event. 8) Counter-clockwise or figure-8 hysteresis may occur during runoff events generated by snowmelt, when channel-margin sediment storage may be protected from gidly-channel flows by snow cover on the rising limb but exposed to flow later in the event. Hypothesis 8 was sustained. Counter-clockwise and figure-8 hystereses in Q-SSC relations were observed at both gully gauging stations during RunoffEvent 11, and at Station B during RunoffEvent 13, most likely due to snowmelt processes occurring throughout the events. 9) Within runoff events in gidlies, duration of suspended sediment transport is brief relative to the duration of runoff events. Suspended sediment transport in gully-channels is almost always supply-limited. Hypothesis 9 was sustained. Runoff-event duration at the gully gauging stations was typically two to five days, while duration of SSC above the moderate level of 20 mg/1 was typically only 3 to 12 hours during the few runoff events in which SSC exceeded 20 mg/1 at all. Once again, RunoffEvent 13 at Station C stands out as anomalous due to the sustained period of moderately high SSC (20 mg/1 < SSC < 100 mg/1). In general, suspended sediment transport usually occurred during brief pulses of a few hours duration within runoff events; SSC pulse duration was usually less than 10% as long as runoff event duration 175 7.1.4 Suspended sediment transport in streams : seasonal and sub-seasonal patterns 10) Suspended sediment transport in gravel-bed streams occurs almost entirely during large rainstorm and/or snowmelt runoff events. Hypothesis 10 was sustained. At Russell Creek Gauging Station, 99% of recorded suspended sediment transport occurred during runoff events. 11) Regardless of antecedent hydrometeorological conditions, the largest runoff events transport the greatest quantities of suspended sediment. Hypothesis 11 was sustained. At Russell Creek Gauging Station, 94% of recorded suspended sediment transport occurred during Runoff Events 6 and 11, which had the highest peak discharges of the study period. 12) The influence of antecedent hydrometeorological conditions on small and intermediate runoff-event suspended sediment transport is weaker in streams than in gullies due to variable conditions over larger drainage area and the buffering effect of in-channel sediment storage. Hypothesis 12 was partially sustained: During the portion of the study period in which Russell Creek SSC data were collected, suspended sediment transport was strongly controlled by discharge characteristics. The sediment supply regimes identified in the gullies were not evident in Russell Creek. By example, during the first runoff event of the wet season, suspended sediment transport in Russell Creek was not abnormally high. However, snowmelt processes proved to be the most important sediment-supply replenishment agent in the study gullies. Since Russell Creek SSC data are missing during the snowmelt period, effect of snowmelt on Russell Creek sediment supply cannot be evaluated. 176 7.1.5 Suspended sediment transport in streams : synoptic scale patterns 13) During small to intermediate runoff events, Q-SSC relations exhibit weak clockwise hysteresis. During larger runoff events in which armour-layer entrainment occurs, however, clockwise hysteresis in Q-SSC relations may occur. Hypothesis 13 was partially sustained. During small to intermediate runoff events, suspended sediment transport in Russell Creek occurred in very brief rising-limb pulses of usually less than 20 mg/1 SSC; total amount transported during these pulses was negligible. During the larger Runoff Events 6 and 11, data limitations preclude clear assessment of Q-SSC relations and sediment sources; however, both bed material and hillslopes/gullies appear to have been important fine-sediment contributors during these events. 14) Suspended sediment transport during runoff events in gravel-bed streams exceeds the duration in gtdly channels, but is also limited in duration by supply rather than by runoff duration. Hypothesis 14 was partially sustained. Suspended sediment transport during runoff events in Russell Creek appeared to occur during discrete high-SSC episodes of much shorter duration than runoff events. Somewhat surprisingly, runoff event durations and high-SSC pulse durations did not differ greatly between the study gullies and Russell Creek. However, SSC data truncation during Runoff Events 6 and 11 prevents the drawing of strong conclusions. 177 7.1.6 Debris-torrent versus fluvial fine-sediment transport 15) Long-term fluvial sediment yield from torrent-prone gidlies approaches debris-torrent yield when only the fine sediment fraction is considered. Hypothesis 15 was tentatively sustained. Debris torrenting and fluvial suspension may be of approximately equal importance in transporting fine sediment out of Gully B. This conclusion is tentative due to the problems of assessing study-period representativeness. 7.1.7 Summary: temporal characteristics of normal-regime fine-sediment transfers Discussion of study findings above clearly illustrates that normal-regime transfer of fine sediment through the hillslope-gully-stream cascade pathway does not occur at a steady, "background" rate. Rather, normal-regime transfer consists of individual, spatially and temporally discrete transfer episodes. Sediment transfers from hillslope sources to gully channel-margin storage sites were linked to discrete synoptic and sub-seasonal scale hydrometeorological conditions. Suspended sediment transport in the study gullies and in Russell Creek occurred in well-defined episodes; episode magnitudes were largely explained by hydrometeorological conditions and antecedent sediment-supply conditions. Clearly, the separation of "episodic events" ~ such as debris torrents — from "normal-regime processes" ~ such as fluvial sediment transport in gullies and gravel-bed streams ~ is practical, but may mask the episodic nature of normal-regime geomorphic activity. 7.2 Application to suspended sediment sampling programs The episodic nature of normal-regime fine-sediment transfers is of general importance to the design of suspended sediment sampling programs. The specific results of this study are 178 applicable to suspended sediment sampling in small coastal British Columbia drainage basins, such as the Tsitika Watershed Sedimentation Study (Hogan and Chatwin 1990). Variability in suspended sediment transport at seasonal, sub-seasonal, and synoptic time scales can greatly affect the representativeness of sediment samples collected. The Tsitika Watershed Sedimentation Study was designed to determine the relative importance of natural versus logging-related fine-sediment transfers into the Tsitika Watershed drainage network. The study consisted of two suspended sediment sampling components. Manual samples of ditch and gully flow downstream of various sediment sources were collected during runoff events to represent fine-sediment production to the drainage network by similar sediment sources throughout the 400-km2 watershed. Suspended sediment transport records from the WSC Gauging Stations on Tsitika River, Russell Creek, and Catherine Creek are to be used as sediment-budget checks. The synoptic-scale variability in suspended sediment output from gullies makes manual sampling of gully flow very difficult. The majority of transport occurs near the beginning of the runoff event, and SSC generally occurs in brief pulses. Therefore, manual sampling must begin at the onset of runoff events — a difficult task in remote locations -- and sampling must occur frequently throughout runoff events ~ an impracticality if many sites are to be sampled. The seasonal and sub-seasonal variability in suspended sediment output from gullies also poses difficulties by complicating runoff-transport relations. Sampling during only either "replenished-supply" or "depleted-supply" regimes will lead to misleading transport results. More likely, sampling during both sediment supply regimes will obscure any relation, leading to inconclusive results. To solve these problems requires development of a qualitative, hydrometerologically-based model of fine-sediment transfer activity. Such a model would serve two main purposes. 179 First, it would aid in interpretation of sediment sampling results. Second, the model would allow rudimentary forecasting of major transfer episodes, based on regional weather forecasts, and allow special sampling efforts to be focussed during these episodes. Elements of a qualitative model of fine-sediment transfer activity are provided in the next section. 7.3 Elements of a normal-regime fine-sediment transfer activity model A qualitative model of normal-regime fine-sediment transfer activity must include the following elements: 1) link between regional weather forecast station and local hydrometeorological conditions; 2) inventory of hillslope sediment sources connected to drainage network; 3) knowledge of sediment-transfer processes likely to operate at hillslope sediment source types and relation of processes to hydrometeorological conditions; 4) nature of sediment storage along low-order channels, especially in gullies where hillslope sediment-transfer processes are concentrated; 5) relation of hydrometeorological conditions to runoff generation; 6) relation of runoff characteristics and sediment supply characteristics to suspended sediment transport in channels of various sizes. In this study, all model elements were considered, but spatial and temporal extent of the study was too restricted to confidently build a complete model. As an example, however, the study results are presented below in the form of the listed model elements. 180 1) Hydrometeorological conditions in Russell Creek Basin compared to Port Hardy Airport: Rainfall: R R 2 = 1.17 R P H Minimum temperature: Min. TR2 = (0.92 x Min. T P H ) - 2.2 Maximum temperature: Max. TR2 = (1.80 x Max. T P H ) - 7.0 (Notation defined in Section 4.1.2). The relation between Russell Creek Basin snowpack characteristics and Port Hardy Airport weather conditions was not determined: Given the importance of snowmelt-related sediment-transfer processes, modelling snowpack conditions is important. 2) Inventory of sediment sources in the study gullies: Gully B: 8 sidewall failures, 4 active sidewall slopes, 4 road crossings, 1 large LOD-jam sediment wedge, numerous smaller wedges; Gully C: undercut banks, small LOD-jam sediment wedges. 3) Hillslope to gully-channel sediment-transfer processes and hydrometeorological controls: Sidewall failures: Sheetwash and shallow sediment-veneer slides presumably occur during heavy rainfall and/or snowmelt; Active sidewall slopes and road cutbanks: Dry ravel, rainsplash, and frost action presumably occur during dry season, early wet season, and winter cold spells, respectively. Banks and wedge surfaces: Bank undercutting and sediment entrainment from wedge surfaces presumably occur during high gully-channel discharge.. In both gullies, various snowmelt-related processes appear to be the most important agent in transferring sediment into gully channels. 181 4) Channel-margin sediment-storage characteristics: Gully B: channel lined along much of its length by sediment-storage compartments (aprons and cones); these act as buffers between hillslope sediment inputs to gully and fluvial sediment transport down-gully; Gully C: negligible channel-margin sediment storage, unbuffered system. 5) Hydrometeorological conditions and runoff: The discharge threshold in Gully B ~ 270 l/s — provides a useful reference flow. During the study period, the threshold discharge was marginally exceeded four times and greatly exceeded twice. Russell Creek Basin hydrometeorological conditions which resulted in marginal threshold exceedence were two-day rainfall totals of 129 mm on no snowpack (Event 2), 102 mm on a shallow snowpack (Event 7), and about 50 mm on shallow snowpack following a prolonged period of radiation snowmelt (Event 15). Event 13 was generated entirely by radiation snowmelt. These results highlight the importance of snow accumulation and ablation modelling in predicting important runoff events. Looking at rainfall alone, however, all rainfall episodes in which two-day totals exceeded 100 mm resulted in Station B discharges above 270 l/s (Events 2, 6 and 7). 6) Relating runoff characteristics and sediment supply characteristics to suspended sediment transport in gully and stream channels: Sediment supply regimes: Gully B: - unlimited sediment supply occurs when peak discharge exceeds 270 l/s, thus can be treated as replenished supply, although depletion occurs during large 182 events; - replenishment of sub-threshold sediment supply occurs during snowmelt, following winter cold spells and perhaps following the dry season; these replenishments have little effect on total event transport in this buffered system. Gully C: - replenished sediment supply occurs following dry season and during snowmelt period; replenishments have significant effect on total event transport in unbuffered system. Russell Creek: no sediment supply regimes identified. Regime-specific flow-transport relations: Gullies, replenished-supply regime: T = 9.0 V 0 9 5 (kg, 103 m3) Gullies, depleted-supply regime: T = 1.1 V 0 6 9 (kg, 103 m3) Russell Creek: T = 0.8 V 3 5 7 (103 kg, 106 m3) where T = suspended sediment transport and V = runoff volume, for individual independent runoff events. 7.4 Conclusions This study has shown that normal-regime fine-sediment transfer activity in a small coastal British Columbia drainage basin occurs in the form of well-defined, discrete, episodic events at specific sites which occupy a small fraction of the landscape. Normal-regime sediment transfers are typically considered very differently from debris torrents and landslides because smaller magnitudes and higher spatial and temporal frequencies are involved. However, the episodic nature of normal-regime sediment transfers suggests that intensive event-focussed sampling is of greater use than low-intensity long-duration sampling in understanding and quantifying normal-183 regime geomorphic activity. The disctrete nature of sediment transfers is also encouraging to initiatives such as the current provincial Watershed Restoration Program; since sediment transfers occur at discrete, identifiable sediment sources, alleviation of logging-induced sediment production should be possible. In other words, rehabilitation of the worst erosion sites should provide beneficial results. The elements of a qualitative normal-regime sediment-transfer model have been provided. To build a useful model upon these foundations, sampling should be extended spatially to cover more sediment-source and gully types, as well as replicates within each type. Sampling should also be extended temporally to cover a greater range and number of combinations of hydrometeorological conditions. Relations between local and regional hydrometeorological conditions require considerable refinement, especially with respect to snow accumulation and ablation. Most importantly, research effort should be focussed on understanding sediment transfers at the runoff-event scale. 184 REFERENCES Anderson, B.A. and D.F. Potts. 1987. Suspended sediment and turbidity following road construction and logging in western Montana. Water Resour. Bull., 23: 681-690. i • • • • Anderson, E.W. and N.J. Cox. 1978. A comparison of different instruments for measuring soil creep. Catena, 5: 81-93. Barr, D.J. and D.N. Swanston. 1970. Measurement of creep in a shallow, slide-prone till soil. Amer. Jour, of Sci., 269: 467-480. Beschta, R.L. 1978. Long-term patterns of sediment production following road construction and logging in the Oregon Coast Range. Water Resour. Res., 14: 1011-1016. Beschta, R. L. 1980. Turbidity and suspended sediment relationships. Symposium on Watershed Management, ASCE, Irrigation and Drainage Division, Boise, Idaho. Vol. I: 271-282. Beschta, R.L. and W.L. Jackson. 1979. The intrusion of fine sediments into a stable gravel bed. Jour. Fish. Res. Board of Can., 36: 204-210. Bilby, R.E., K. Sullivan, and S.H. Duncan. 1989. The generation and fate of road-surface sediment in forested watersheds in southwestern Washington. For. Sci., 35: 453-468. Bjornn, T.C. and D.W. Reiser. 1991. Habitat requirements of salmonids in streams. In. Meehan, W.R. (ed.). Influences of Forest and Rangeland Management on Salmonid Fishes and Their Habitats. American Fisheries Society Special Publication 19, Bethesda, Maryland, p. 83-138. Bley, D. and K.H. Schmidt. 1991. Forschungen zur Fluss- und Hangdynamik. Freiburger Geographische Hefte, 33: 121-129. Bovis, M.J. and B.R. Dagg. 1987. Mechanisms of debris supply to steep channels along Howe Sound, southwest British Columbia. //;: Erosion and Sedimentation in the Pacific Rim. IAHSPubl. 165. p. 191-200. British Columbia Ministry of Environment, Lands and Parks and Ministry of Forests. 1994. Watershed Assessment Procedure (Interim Methods). Watershed Restoration Technical Circular No. 2, Victoria, B.C. 63 pp. British Columbia Ministry of Forests. 1992. Biogeoclimatic Zones of British Columbia Map, Victoria, B.C. Caine, N. and F.J. Swanson. 1989. Geomorphic coupling of hillslope and channel systems in two small mountain basins. Zeit. Geomorph., 33: 189-203. Carson, M.A. and M.J. Kirkby. 1972. Hillslope Form and Process. Cambridge University Press, London. 475 pp. 185 Cheng, J. D., T.A. Black, and R.P. Willington. 1975! The generation of stormflow from small forested water-sheds in Coast Mountains of southwestern British Columbia. In: N.R.C. Hydrology Symposium Proc, Winnipeg, Man. p. 544-551. Church, M . 1983. Concepts of sediment transfer and transport on the Queen Charlotte Islands. Fish/Forestry Interaction Program, Working Paper 2/83, Victoria, B.C. 30 pp. Church, M . .1992. Channel morphology,and typology. In: Calow, P. and G.E. Petts (eds.). The Rivers Handbook: Hydrological and Ecological Principles. Blackwell Sci. Publ. 1, Oxford, U.K. p. 126-143. Church, M . In press. Fine sediments in small streams in coastal British Columbia: a review of research progress. In: Hogan, D.L. and P. Tschaplinsky (eds.). Proc. of the Fish/Forestry Workshop on Carnation Creek and FFIP, Queen Charlotte City, B.C., May 1994. Church, M . and R. Kellerhals. 1970. Stream Gauging Techniques for Remote Areas Using Portable Equipment. Dept. of Energy, Mines, and Resources, Inland Waters Branch, Tech. Bull. No. 25. Ottawa, Ont. 90 pp. Dietrich, W.E., T. Dunne, N.F. Humphrey and L . M . Reid. 1982. In: Swanson, F.J., R.J. Janda, T. Dunne, and D.N. Swanston (eds.). Sediment Budgets and Routing in Forested Drainage Basins. USDA For. Serv. Gen. Tech. Rep., PNW-141. Portland, Ore. p. 5-23. Dietrich, W.E., S.L. Reneau and C.J. Wilson. 1987. Overview: "zero-order basins" and problems of drainage density, sediment transport and hillslope morphology. In: Erosion and Sedimentation in the Pacific Rim. IAHS Publ. No. 165. p. 27-37. Doty, R.D., H.B. Wood, and R.A. Merriam. 1981. Suspended sediment production from forested watersheds on Oahu, Hawaii, Water Resour. Bull., 17: 399-405. Duncan, S.H., R E . Bilby, TW. Ward, and J.T. Heffner. 1987. Transport of road-surface sediment through ephemeral stream channels. Water Resour. Bull., 23: 113-119. Environment Canada, Atmospheric Environment Service. 1993. Canadian Climate Normals, 1961-90, British Columbia. Ottawa, Ont. 134 pp! Environment Canada, Atmospheric Environment Service. 1995. Surface weather synopsis maps. Environment Canada Data Services, Vancouver, B.C. Everest, F.H., R.L. Beschta, J.C. Scrivener, K.V. Koski, J.R. Sedell, and C.J. Cederholm. 1987. Fine sediment and salmonid production: a paradox. In: Salo, E.O. and T.W. Cundy (eds.), Streamside Management:' Forestry and Fishery Interactions. The University of Washington, Institute of Forest Resources Contribution 57, Seattle, Wash. p. 98-142. Frostick, L . E . , P.M. Lucas, and I. Reid. 1984. The infiltration of fine matrices into coarse-grained alluvial sediments and its implications for stratigraphical interpretation. Jour. Geol. Soc. of London, 141: 955-965. 186 Geological Survey of Canada. 1983. Map 1552A : Geology, Alert Bay - Cape Scott, British Columbia. Ottawa, Ont. Gibbons, D.R. and E O . Salo. 1973. An annotated bibliography of the effects of logging on fish of the western United States and Canada. U.S. For. Serv. Gen. Tech. Rep., PNW-10. Portland, Ore. 139 pp. Golding, D L . 1987. Changes in streamflow peaks following timber harvest of a coastal British Columbia watershed. IAHS Publ. 167. p. 509-517. Heede, B.H. 1985. Interaction between streamside vegetation and stream dynamics. In: Proc. of the Symposium, Riparian Ecosystems and their Management: Reconciling Conflicting Uses. U.S. For. Serv., Region 3, Albuquerque, N.M. p. 54-58. Henderson-Sellers, A. and P.J. Robinson. 1986. Contemporary Climatology. Longman Scientific and Technical, Harlow, U.K. 439 pp. Hetherington, E.D. 1982. Effects of forest harvesting on the hydrologic regime of Carnation Creek Experimental Watershed: a preliminary assessment. Proceedings, Canadian Hydrology Symposium: 82, Fredericton, N.B. p. 247-267. Hicks, B.J., J.D. Hall, P.A. Bisson, and J.R. Sedell. 1991. Responses of salmonids to habitat changes. In: Meehan, W.R. (ed.). Influences of Forest and Rangeland Management on Salmonid Fishes and Their Habitats. American Fisheries Society Special Publication 19, Bethesda, Maryland, p. 483-518. Hogan, D.L. 1987. The influence of large organic debris on channel recovery in the Queen Charlotte Islands, British Columbia, Canada. In: Erosion and Sedimentation in the Pacific Rim.'IAHS Publ. 165. p. 343-353. Hogan, D.L. and S. Chatwin. 1990. Tsitika River Sedimentation Monitoring Program: Draft Working Plan. B.C. Ministry of Forests, Victoria, B.C. 17 pp. Holstrom, T.A. and R.H. Hawkins. 1980. Particle size distribution effects on turbidity. Symposium on Watershed Management, ASCE, Irr. and Drain. Div., Boise, Idaho. Vol. I: 283-297. Hudson, R. 1995. Hydrologic recovery due to snow accumulation and melt at Upper Gray Creek, Sunshine Coast: preliminary analysis. In: Guy, B.T. and J. Barnard (eds.). Mountain Hydrology: Peaks and Valleys in Research and Applications. Canadian Water Resources Assoc., Cambridge, Ont. p. 23-28. Hungr, O., G.C. Morgan and R. Kellerhals. 1984. Quantitative analysis of debris torrent hazards for design of remedial measures. Can. Geotech. Jour., 21: 663-677. Jackson, W.L. and R.L. Beschta. 1982. A model of two-phase bedload transport in an Oregon Coast Range stream. Earth Surf. Proc. and Land., 7: 517-527. Klinka, K., R.N. Green, P.J. Courtin and F.C. Nuszdorfer. 1984. Site diagnosis, tree species selection, and slashburning guidelines for the Vancouver Forest Region, British Columbia. B.C. Min. of Forests, Land Man. Report 25, Victoria, B.C. 180 pp. 187 Lawler, D M . 1988. Environmental limits of needle ice: a global survey. Arctic and Alpine Research, 20: 137-159. Lawler, D.M. 1992. Design and installation of a novel automatic erosion monitoring system. Earth Surf. Proc. and Land., 17: 455-463. Lawler, D.M. 1993. Needle ice processes and sediment mobilization on river banks: the River Ilston, West Glamorgan, UK. Jour, of Hydrol. (Neth.), 150: 81-114. Lehre, A. K. 1982. Sediment budget of a small Coast Range drainage basin in north-central California. In: Swanson, F.J., R.J. Janda, T. Dunne, and D.N. Swanston (eds). Sediment Budgets and Routing in Forested Drainage Basins. USDA For. Serv. Gen. Tech. Rep., PNW-141. Portland, Ore. p. 67-77. Likens, G.E. and R.E. Bilby. 1982. Development, maintenance, and role of organic-debris dams in New England streams. In: Swanson, F.J., R.J. Janda, T. Dunne, and D.N. Swanston (eds.). Sediment Budgets and Routing in Forested Drainage Basins. USDA For. Serv. Gen. Tech. Rep., PNW-141. Portland, Ore. p. 122-128. Linsley, R.K. and J.B. Franzini. 1979. Water-Resources Engineering. McGraw-Hill Book Co., New York. 716 pp. Lopes, V. L. and P.F. Ffolliott. 1993. Sediment rating curves for a clearcut ponderosa pine watershed in northern Arizona. Water Resour. Bull., 29: 369-382. Mackay, J.R. and W.H. Mathews. 1974. Movement of sorted stripes, The Cinder Cone, Garibaldi Park, B.C., Canada. Arctic and Alpine Research, 6: 347-359. Maynard, D. 1991. Tsitika River Watershed: Sediment Source Inventory. B.C. Ministry of Forests, Victoria, B.C. 18 pp. Meehan, W.R. and D.N. Swanston. 1977. Effects of gravel morphology on fine sediment accumulation and survival of incubating salmon eggs. U.S. For. Serv. Res. Pap., PNW-220. Portland, Ore. 16 pp. Megahan, W.F. 1983. Hydrologic effects of clearcutting and wildfire on steep granitic slopes in Idaho. Water Resour. Res., 19: 811-819. Millard, T.H. 1993. Sediment in Forested and Logged Gullies, Coastal British Columbia. M.Sc. thesis, The University of British Columbia, Vancouver, B.C. 202 pp. Mosley, M.P. 1981. The influence.of organic debris on channel morphology and bedload transport in a New Zealand forest stream. Earth Surf. Proc. and Land., 6: 571-579. Nanson, G.C. 1974. Bedload and suspended-load transport in a small, steep, mountain stream. Amer. Jour. Sci., 274: 471-486. 188 Oden. M.E. 1994. Debris Recharge Rates in Torrented Gullies on the Queen Charlotte Islands. M.Sc. thesis, The University of British Columbia, Vancouver, B.C. 100 pp. Oke, T. and J. Hay. 1994. The Climate of Vancouver. British Columbia Geographical Series, No. 50, The University of British Columbia, Vancouver, B.C. 84 pp. Olive, L.J. and W.A. Rieger. 1985. Variation in suspended sediment concentration during storms in five small catchments in southeast New South Wales. Australian Geographical Studies, 23:38-51. O'Loughlin, C.L. , L.K. Rowe, and A.J. Pearce. 1978. Sediment yields from small forested catchments North Westland - Nelson, New Zealand. Jour, of Hydrol. (New Zeal.), 17: 1-15. Olyphant, G. A., C P . Carlson, and D. Harper. 1991. Seasonal and storm-related aspects of sediment yield from a rapidly eroding coal refuse deposit in southwestern Indiana. Water Resour. Res., 27: 2825-2833. Outcalt, S.I. 1971. An algorithm for needle ice growth. Water Resour. Res., 7: 394-400. Paustian, S.J. and R.L. Beschta. 1979. The suspended sediment regime of an Oregon Coast Range stream. Water Resour. Bull., 15: 144-154. Piest, R.F., J.M. Bradford, and G M . Wyatt. 1975. Soil erosion and sediment transport from gullies. Jour, of the Hydr. Div. of the Amer. Soc. of Civ. Eng., 101: 65-80. Reid, I. and L . E . Frostick. 1984. Particle interaction and its effect on the thresholds of initial and final bedload motion in coarse alluvial channels. In: Koster, E.H. and R.J. Steel (eds.). Sedimentology of Gravels and Conglomerates. Can. Soc. of Petr. Geol., Memoir 10. p. 61-68. Reid, L . M . 1981. Sediment Production from Gravel-Surfaced Forest Roads, Clearwater Basin, Washington. M.Sc. thesis, The University of Washington, Seattle, Wash. 247 pp. Reid, L . M . and T. Dunne. 1984. Sediment production from forest road surfaces. Water Resour. Res., 20: 1753-1761. Reid, L . M . , T. Dunne, and C.J. Cederholm. 1981. Application of sediment budget studies to the evaluation of logging road impact. Jour, of Hydrol. (New Zeal.), 20: 49-62. Roberts, R.G. and M . Church. 1986. The sediment budget in severely disturbed watersheds, Queen Charlotte Ranges, British Columbia. Can. Jour. For. Res., 16: 1092-1106. Rood, K . M . 1984. An aerial photograph inventory of the frequency and yield of mass wasting on the Queen Charlotte Islands, British Columbia. B.C. Min. of Forests, Land Man. Report 34, Victoria, B.C. 55 pp. Schumm, S.A. and R.W. Lichty. 1965. Time, space, and causality in geomorphology. Amer. Jour, of Sci., 263: 110-119. 189 Scrivener, J.C. and M.J. Brownlee. 1989. Effects of forest harvesting on spawning gravel and incubation survival of chum and coho salmon in Carnation Creek, B.C. Can. Jour, of Fish, and Aqu. Sci., 37: 681-696. Selby, M.J. 1985. Earth's Changing Surface. Clarendon Press, Oxford, U.K. 607 pp. Sidle, R.C., A.J. Peace and C.L. O'Loughlin. 1985. Hillslope Stability and Landuse. Am. Geophys. Union, Washington, D.C. 140 pp. Sidle, R.C. 1988. Bed load transport regime of a small forest stream. Water Resour. Res., 24: 207-218. Sidle, R.C. and A.J. Campbell. 1985. Patterns of suspended sediment transport in a coastal Alaska stream. Water Resour. Bull., 21: 909-917. Smith, L .C. and G.A. Olyphant. 1994. Within-storm variations in runoff and sediment export from a rapidly eroding coal-refuse deposit. Earth Surf Proc. and Land., 19: 369-375. Smith, R.D., R.C. Sidle, and P.E. Porter. 1993. Effects on bedload transport of experimental removal of woody debris from a forest gravel-bed stream. Earth Surf. Proc. and Land., 18: 455-468. Strahler, A.N. 1957. Quantitative analysis of watershed geomorphology. Trans. Amer. Geophys. Union, 38: 913-920. Suckling, P.W. 1977. A Solar Radiation Model and an Analysis of Synoptic Solar Radiation Regimes in British Columbia. Ph.D. Thesis, The Univeristy of British Columbia, Vancouver, B.C. 271 pp. Swanson, F.J., L . E . Benda, S.H. Duncan, G E . Grant, W.F. Megahan, L . M . Reid, and R.R. Ziemer. 1987. Mass failures and other processes of sediment production in Pacific Northwest forest landscapes. In: Salo, E.O. and T.W. Cundy (eds.). Streamside Management: Forestry and Fishery Interactions. College of For. Resour., The University of Washington, Seattle, p. 9-38. Swanson, F.J. and C T . Dyrness. 1975. Impact of clearcutting and road construction on soil erosion by landslides in the western Cascade Range, Oregon. Geology, 3: 393-396. Swanston, D.N. 1981. Creep and earthflow erosion from undisturbed and management impacted slopes in the Coast and Cascade Ranges of the Pacific Northwest, U.S.A. In: Davies, T.R.H. and A.J. Pearce (eds.). Erosion and Sediment Transport in Pacific Rim Steeplands. IAHSPubl. No. 132. p. 76-94. Swanston, D.N. and F.J. Swanson. 1976. Timber harvesting, mass erosion, and steepland forest geomorphology in the Pacific Northwest. In: Coates, D.R. (ed.). Geomorphology and Engineering. Dowden, Hutchison and Ross, Stroudsburg, Pa. p. 199-221. 190 Swanston, D.N. 1991. Natural processes. //;: Meehan, W.R. (ed.). Influences of Forest and Rangeland Management on Salmonid Fishes and Their Habitats. American Fisheries Society Special Publication 19, Bethesda, Maryland, p. 139-179. Tassone, B.L. 1987. Sediment loads from 1973 to 1984, 08HB048 Carnation Creek at the Mouth, British Columbia. In: Chamberlin, T. W. (ed.). Proc. of the Workshop: Applying 15 Years of Carnation Creek Results. Carnation Creek Steering Committee, Nanaimo, B.C. p. 46-58 Toy, T.J. 1983. A comparison of the LEMI and erosion pin techniques. Zeit. Geomorph., Suppl.-Bd. 46: 25-34. Tripp, D.B. and V.A. Poulin. 1986. The effects of logging and mass wasting on salmonid spawning habitat in streams on the Queen Charlotte Islands. B.C. Min. of Forests, Land Man. Report 50, Victoria, B.C. 29 pp. Tropeano, D. 1991. High flow events and sediment transport in small streams in the 'Tertiary Basin' area in Piedmont (northwest Italy). Earth Surf. Proc. and Land., 16: 323-339. Truhlar, J.F. 1978. Determining suspended sediment loads from turbidity records. Hydrol. Sci. Bull., 23:409-417. Tuckfield, C G . 1964. Gully erosion in the New Forest, Hampshire. Amer. Jour, of Sci., 262: 795-807. Van Sickle, J. and R.L. Beschta. 1983. Supply-based models of suspended sediment transport in streams. Water Resour. Res., 19: 768-778. VanDine, D.F. 1985. Debris flows and debris torrents in the Southern Canadian Cordillera. Can. Geotech. Jour., 22: 44-68. Williams, G.P. 1989. Sediment concentration versus water discharge during single hydrologic events in rivers. Jour, of Hydrol. (Neth.), 111: 89-106. Wolman, M.G. and J.P. Miller. 1960. Magnitude and frequency of forces in geomorphic processes. Jour. pfGeol., 68: 54-74. Wood, P. A. 1977. Controls of variation in suspended sediment concentration in the River Rother, West Sussex, England. Sedimentology, 24: 437-445. Wright, K.A. , K.H. Sendek, R M . Rice and R B . Thomas. 1990. Logging effects on streamflow: storm runoff at Caspar Creek in northwestern California. Water Resour. Res., 26: 1657-1667. Wu, T.H. , W.P. McKinnell, and D.N. Swanston. 1979. Strength of tree roots and landslides on Prince of Wales Island, Alaska. Can. Geotech. Jour., 16: 19-33. 191 OD O in N - 3 3 73 • o w "° o —i • oi u eft oi yi -» oi o oi Ol o o 0 1 Z Z 2 "2 1 ^* co to o w O I m m co Ct U O l tv) -J> CO m i o> 1 bi x1 H us a* V3 rt > n <' rt o •a rt C 0 0 1 O 1 S A W W - * c n o > c o o o m o o r t ^ O O M O U N - ' l o o ! cn | io 1 1 8 8 S K 1 0 0 ) - k O > - k O O . O . CD CD rO -%J -«4 =r: =r. — (O CO N) O. Q. Q. <D CD CD i i i i i i I H p cr Si oo 11 c •1 !? I tn cn t/i o o o 3 3 3 3 o tn C* 3 CD 0) O Q. - a | CO & g =" £ 3' cn cn o o co 3 ro o • CD 00 C L CQ 0J CD CT Si Si 3 cr J O C l OJ 3" Q. 5 -Q. cn cs 31 0 CO c_ •< o §• 3 1 8 3 3 CD CQ CO cr o a tn cs Q. 3 a to CB 3 r co <g cn — E. 2 IQ •< CD ^2. — CB 5 CD 3 M * 5 ! <2 CD Q. 01 5= = 01 0 s-S. £ CO £  0 D "5 ' 3 01 cr o < CD 3 cu Q. H CO ro o O ro fr o oi r i "i p co o> — P CT O CD u o n> 3 a Q. -» 8. S <2 3 o ro H a cr C/5 rt a 0 > •a •a S D. w o c/> o' s 5 •a o 5 5' a co CB a CD Q. CD < CB O CO CD "—' CO TJ 3' cn V ro Bo cn to 3 o < CD a o 3 ~j c_ CU 3 CD CT CO TJ OJ rO I cn o 3 3 CB 3 CQ II co co z CD * ' CD N> co co ^ Z — 3 " • * f t CB' a o cu Q. CB < < s i a a CB  Q. 8 * < CD 3 D . O & o. CB < CB O •o 3 CD z CD CD Q. CD 3 £ Ol D CD O CB X 8 to 00 O 3 3 "tj 3! 5 ' 1 3 • w » •j* 1 OJ t = -» 3. 3. O O -L "> ~ J O c_ - 3 £ co £ S » f 8 ? € " "O 5 ' Ol Ul A W M -•• to M M CO Ol Ol O) Q Q 03 CD OJ OJ to M i» «> ^ Ol «-~V r^  V S S z fo K » CO 01 OJ to 1^4 -* xn .o-tn-K , K J to Ol OJ to ^ K - OJ OJ to M Q n cn u to 00 M g <» CO ~J x1 cr rt" o C L cr 3 TJ OJ 10 3' *t 00 Nov. J Pin •00 to 4^ cn -j 0 00 CO OJ Len Vo O 06 Dec. j CQ 5 1. ~ 0.3 rerflowi filled. Screer 06 Dec. j (mm) 3 OJ 3 CO CO OJ cn OJ to -» W Ol A A J« cn 01 o> CO t j ~~ ~~ O l cn 01 fo O l N | >l A to Ol OJ o to x1 H a cr • rt" > C/3 rt U "3 to H cr n" ->... »—' ON C/3 rt c/i n o cr » a O o O l * - OJ cn CB H u cr ro OJ Q to co CB 00 CO tn OJ O l to -1 o to 3' to OJ - J CD O O i u cn A 00 to 00 o 01 to ro 01 OJ O l CD o ro OJ cn to OJ OJ OJ CO o ro OJ OJ 01 co U Ul A w w s 00 O Oi OJ Ol J>. OJ Ol o U Ol A U l A S S"1 CD O o o o. :§' rt 3 as rt O . CM rt CB CTI I Appendix A-2 : Discharge and Suspended Sediment Concentration at Gauging Station B a) Discharge and SSC at Station B, 30 September - 10 October, 1993 10OO 100 10 — * ' t—— 1 • 1 1 i M (C I 1. SSC) •—H-»— M (SSC| j i i I , — f - _._ i i i i i i A Sap 01-Oct 02-Oct 03-Oct i i 0+Oct OSOtt Date i — O&Oct 1 = 1 • 07-Oct O K M 0»Oc* b) Discharge and SSC at Station B, 20 - 30 October, 1993 o , — . • • M (Q SSC) , 1 ! ' ™— —. — . , . _ - — i — i , j I ... A • ~ ~ ~-n;—35—• „. — 1 — t 1 — 1 — — 1 — t — 1 i 1 1 — i — > f~l\ ^ e CB E 2SOd Date Figure A-2.1. Discharge (Q) and suspended sediment concentration (SSC) at Gauging Station B. a) 30 September -10 October, 1993; b) 20 - 30 October, 1993. No data 10-20 October. 193 a) Discharge and SSC at Station B, 30 October - 9 November, 1993 b) Discharge and SSC at Station B, 9-19 November, 1993 „ Runoff Event 3 Figure A-2.2. Discharge (Q) and suspended sediment concentration (SSC) at Gauging Station B. a) 30 October - 9 November, 1993; b) 9 - 19 November, 1993. 194 a) Discharge and SSC at Station B, 19 - 29 November, 1993 • SSC b) Discharge and SSC at Station B, 29 November - 9 December, 1993 SSC Figure A-2.3. Discharge (Q) and suspended sediment concentration (SSC) at Gauging Station B. a) 19 - 29 November, 1993; b) 29 November - 9 December, 1993. 195 a) Discharge and SSC at Station B, 9 - 19 December, 1993 5 . o E o b) Discharge and SSC at Station B, 19 - 29 December, 1993 " Runoff Event I SSC Figure A-2.4. Discharge (Q) and suspended sediment concentration (SSC) at Gauging Station B. a) 9 - 19 December, 1993; b) 19 - 29 December, 1993. 196 b) Discharge and SSC at Station B, 8-18 January, 1994 cr , Kunoff fcvent 1 ! I ! i i 1 i 1 —[- —1 X . - - — ^ —' 1 j ! / \ i | I 1 i ! ' 1 ' 1 !' • ! i i fl ft « • i j , i A t , fl HIT OftWan <»>lan ICKIan l U a n 12Jan 13>lan 14Jan 15>lan 16Wan 17^)an lft^an Date Q SSC Figure A-2.5. Discharge (Q) and suspended sediment concentration (SSC) at Gauging Station B. a) 29 December, 1993 - 8 January, 1994; b) 8 - 18 January, 1994. a) Discharge and SSC at Station B, 18 - 28 January, 1994 • ' • i ' i • i • 1 • 1 ! 1 •. : ——r-.' 19-Jin - 2 C W M 21-J«n 220«n 2 W « n 24^Un 2 W M T&Jwn 27J«n 2frJ«n D a t e — Q S S C 1000 ' b) Discharge and SSC at Station B, 28 January - 7 February, 1994 100 -r:rt::::::™~|:r™::zrirrrzzr , : _ : z-l :: ZrE::: :r:r~r-E:il:.-::T-= -1000 —. — -100 I o 10 -1 -SSC (mg/I) . . j SSC (mg/I) . :r — 1 _ — , —=— :—: * zr : SSC (mg/I) . • - i •b ] '1 1 i 1 ' I I I I 1 r 1 ! 1 ! f 2B~J.n 29>J«n MWin 3W«n 01-F«b 02-F«6 O^Fab 04-F«b 0S-F*b 0&F«t> • 07-F D a t e Q S S C Figure A-2.6. Discharge (Q) and suspended sediment concentration (SSC) at Gauging Station B. a) 18-28 January, 1994; b) 28 January - 7 February, 1994. 198 a) Discharge and SSC at Station B, 7 - 17 February, 1994 — J j j i_ . J 4 J i j j 1——— 4 ; i , i ! -— * -. - —, :_, . - A j ! / i I I • 1 ' 1 : 1 : 1 ! 1 : 1 r- i i 1 E ' : I : I ! 1 :  1 ' 1 0 8 - F * 1 0 - F * 11-F«b 12-F.b l l f . 6 14-F.6 15-F.6 IfrFrt D a t e Q S S C 1000 b) Discharge and SSC at Station B, 17 - 27 February, 1994 100 -: - f _ -• r - — j • — E E . -1000 = ~= 1 „ •——•—-»—-—- -—• ? - - = " -100 a 10 -f E o CO w -10 1 :— = : ; -=2 — j i -17-F«b 1ft-F«b 19-F.b 20-F.b ' ^ : l : 1 ! I ! 1 i f 21-F.b 22-F.fc 23-F.b 2 t f r t 25-F.b 26-F.t, 27-F D a t e Q S S C Figure A-2.7. Discharge (Q) and suspended sediment concentration (SSC) at Gauging Station B a) 7 - 17 February, 1994; b) 17 - 27 February, 1994. 199 a) Discharge and SSC at Station B, 27 February - 9 March, 1994 a 27-Fab 2S-Fab . 01 .Mar . 02-Mar 03-Mar 05~Maf 07-Mat Oft-Mar 00-Maf • SSC b) Discharge and SSC at Station B, 9-19 March, 1994 O) E" SSC Figure A-2.8. Discharge (Q) and suspended sediment concentration (SSC) at Gauging Station B. a) 27 February - 9 March, 1994; b) 9 - 19 March, 1994. 200 a) Discharge and SSC at Station B, 19 - 29 March, 1994 j Runof Event 11 ;ZHJ b) Discharge and SSC at Station B, 29 March - 8 April, 1994 a a 2 * M M » M « r 31+«.r 01-AfX 02-A(x 0 W | » O M < » OSAjx O M d 07-Ap( O M ( » Date Figure A-2.9. Discharge (Q) and suspended sediment concentration (SSC) at Gauging Station B. a) 19 - 29 March, 1994; b) 29 March - 8 April, 1994. 201 a) Discharge and SSC at Station B, 8 - 18 April, 1994 b) Discharge and SSC at Station B, 18-28 April, 1994 E, o Date • SSC Figure A-2.10. Discharge (Q) and suspended sediment B. a) 8 - 18 April, 1994; b) 18 - 28 April, 1994. concentration (SSC) at Gauging Station 202 o & a a) Discharge and SSC at Station B, 28 April - 8 May, 1994 "M (<3.SSC)~ • SSC b) Discharge and SSC at Station B, 8 May - 18 May, 1994 M (Q, SSC) ~ ~ M (SSC) «—rH ^ — i - M ,A T A C**Uy OMtay t l M t o y i n t o , 1 M t o , 1 M U y , Date Figure A-2.11. Discharge (Q) and suspended sediment concentration (SSC) at Gauging Station B. a) 28 April - 8 May, 1994; b) 8 - 18 May, 1994. 203 a) Discharge and SSC at Station B, 18-28 May, 1994 o -+—4 H o i * i - i — i — i — i — h i 1 « - * b y 1 9 4 ^ y 20Mn 2 1 - U . y 2 2 - u . y 2 M t o y 2 « U v M - M . , 2 8 - M . , 2 7 * 1 . , 2 W K ^ Date • SSC b) Discharge and SSC at Station B, 28 May - 7 June, 1994 o r—4-E Q ' SSC Figure A-2.12. Discharge (Q) and suspended sediment concentration (SSC) at Gauging Station B. a) 18 - 28 May, 1994; b) 28 May - 7 June, 1994. 204 Discharge and SSC at Station B, 7-17 June, 1994 -Runof f Event 17 H M (Q, SSC) 07Ji*i OfrJun 09-Jun iftjun 11 .Jun 12-Jun - 13-Jun 14Jun Date Figure A-2.13. Discharge (Q) and suspended sediment concentration (SSC) at Gauging Station B, 7 - 17 June, 1994. N o data, 17 - 27 June. 205 Appendix A-3 : Discharge and Suspended Sediment Concentration at Gauging Station C a) Discharge and SSC at Station C, 30 September - 10 October, 1993 o S O ^ p • 01-Ocl 02-Oel C O - o a o t o a o s o c t c e o a 0 7 - o a o x u o a o * 1 0 0 a Data b) Discharge and SSC at Station C, 10 - 20 October, 1993 a CO E 1 « W i v c * tJ-CW l i O c t 140d i s o a ^ 1 4 < w Date Figure A-3.1. Discharge (Q) and suspended sediment concentration (SSC) at Gauging Station C a) 30 September - 10 October, 1993; b) 10 - 20 October, 1993. 206 a) Discharge and SSC at Station C, 20 - 30 October, 1993 & a » o a 2ioa 22-oa j j o a 24-oa - A -2 « « 2S<W 27-Od 28-Od 7*Oa 3 0 0 3 Date b) Discharge and SSC at Station C, 30 October - 9 November, 1993 Figure A-3.2. Discharge (Q) and suspended sediment concentration (SSC) at Gauging Station C. a) 20 - 30 October, 1993; b) 30 October - 9 November, 1993. 207 a) Discharge and SSC at Station C, 9 - 19 November, 1993 5. o ~ Runoff Event 3 i ,_ 09-Nov 1CH<ov ... 11-Nov 12-Nov IW(ov 14-Nov 15-No. 16-Nov 17-Nov 18-Nov 19-N« ' ' '. " D a l e SSC b) Discharge and SSC at Station C, 19 - 2? November, 1993 s Runoff Event 6 o SSC Figure A-3.3. Discharge (Q) and suspended sediment concentration (SSC) at Gauging Station C. a) 9 - 19 November, 1993; b) 19 - 29 November, 1993. 208 Cf a) Discharge and SSC at Station C, 29 November - 9 December, 1993 f E S S C o b) Discharge and SSC at Station C, 8 - 18 January, 1994 j ; , i i ! j M (Q, SSC] i i . 1 — \ i l ! I % E 08~Jan 09-J«n 10-Jsn 1 W a n 12^Jan 13~J«n Date 14-Jin 1S-Jan • , 1kVJ«n 17-Jan ItVJao Figure A-3.4. Discharge (Q) and suspended sediment concentration (SSC) at Gauging Station C. a) 29 November - 9 December, 1993; b) 8 - 18 January, 1994. No data, 9 December - 8 January. 209 1000 -b) I 2 )ischai S Janu -gc and SSC at Stati ary - 7 February, 19 on C, 94 p1000 100 -: j -100 j i . ' ! • o 10 -L i f E o to (0 -10 1 - -1 •b i i ; i | 1 i I • 1 : 1 ! 1 • 1 : 1 : 1 ! 1 ! 1 2&J«n 29^ 111 3(W«n 31>lan 01-f»to 02-F«b 0 3 ^ « b 04-F«b OVfcb 06-F«b 07-f Date — — Q S S C Figure A-3.5. Discharge (Q) and suspended sediment concentration (SSC) at Gauging Station C. a) 18 - 28 January, 1994; b) 28 January - 7 February, 1994. o a) Discharge and SSC at Station C, 17 - 27 February, 1994 ! ! I I . _ ._. i ! j M (aSSC] i i , i too - ! ! 1 H > | i i ; i ! j 10 -| i i j • i ! i i j i j i i „ . , i i i l l ! I i l i ! ; : • I I i i i j 1 - i l l f i l l 17-F.O 18-F.6 I K * 20-F.6 21-F«6 22-F«6 23-F«b 24-F.6 2S-F* 26-f* 27-F.O Date b) Discharge and SSC at Station C, 27 February - 9 March, 1994 Figure A-3.6. Discharge (Q) and suspended sediment concentration (SSC) at Gauging Station C. a) 17 - 27 February, 1994; b) 27 February - 9 March, 1994. No data, 7 - 17 February. 211 b) Discharge and SSC at Station C, 19 - 29 March, 1994 1000 i ; ..j. ; ;.. .. „ • . „ . , ; RunoffEvent 13 • j — — f : j •+ — - f 1—•—-\ 1 • • | ! • ' . , . ! I i j j 1 , — s ! ICOAar 20-Mar 21-M« 22-M.i 23-Mv 24-Mar 2SM*r 2&Mar 27-M«r 284fcf - I»Mif Date Q SSC Figure A-3.7. Discharge (Q) and suspended sediment concentration (SSC) at Gauging Station C. a) 9 - 19 March, 1994; b) 19 - 29 March, 1994. 212 a) Discharge and SSC at Station C, 29 March - 8 April, 1994 I Runoff Event 13 ~ 4 -I Runoff Event 14 ~ 294ter 30-Mw 31-Mir OI^ Apr 07-Apr 03-Apr 04-Apr OS-Apr 06-Apr 07-Apr OS-Apr Date o b) Discharge and SSC at Station C, 8 - 18 April, 1994 i—.— - - i - -4 Runoff Event IS ^ ^ _ Runon tverrt 16 ; j -----i i M (SSC) | ! j H—s 1 i \ I ! ; i i ! |. | 1 • 1 ! • • 1 | / 1 1 1 i * ^ = ; i ! T I | !- ^  i i j j 1 08-Apr OS-Apr 10-Apr 11-Apr 12-Apr 13-Apr 1+Apr 1S-Apr 15-Apr 17-Apr 18-Apr Date Figure A-3.8. Discharge (Q) and suspended sediment concentration (SSC) at Gauging Station C. a) 29 March - 8 April, 1994; b) 8 - 18 April, 1994. 213 a) Discharge and SSC at Station C, 18 - 28 April, 1994 Runon" Event 11 ~ 5 . a l^Apr 19-Apr 20-Apr 21-Apr 22-Apr 23-Apr Date 24-Apr 25-Apr 26-Apf 27-Apr 26-Apr b) Discharge and SSC at Station C, 28 April - 8 May, 1994 a o - M ( S S C ) 2 M o t 2 M p r 30-Apr 01*Uy 02-trUy CtMtay 0**U)f 05*ur, OS-M« 07-Hr, OMtay Date Figure A-3.9. Discharge (Q) and suspended sediment concentration (SSC) at Gauging Station C. a) 18-28 April, 1994; b) 28 April-8 May, 1994. 214 a) Discharge and SSC at Station C, 8 May - 18 May, 1994 1 M (! SC) „ i ! \ | I 1 ! i F 1 i h— 1 — 1 1 —-t— ! 1 i — i — 1 — i — l — i — — i — i — s. a 09-Miy 0 » 4 U y • 104UV 11-May 12-May 13-May 144toy 154fey ICOtay 17-Mav Date S S C b) Discharge and SSC at Station C, 7 - 17 June, 1994 I ft Evert 1 7 — i J Runo i=~7T.' i I ! 1 ! i ; i I • ! I | | | i j j 1 I I I ! — i — i — i — i — h — i — i — , 1 P*H i 1 J 1 i 1 1 07-J\m OS^Jun 09~Jun 1&Jun ll^Jun 12^Jun 13Jun 14-Jun 1S>lun 1C>Jun 17-Jun Date Q S S C Figure A-3.10. Discharge (Q) and suspended sediment concentration (SSC) at Gauging Station C. a) 8 - 18 May, 1994; b) 7 - 17 June, 1994. Discharge < 1 l/s, SSC < 1 mg/l, 18 May - 7 June. 215 Discharge and SSC at Station C, 17 - 27 June, 1994 4 ^ - P — i h H 1 h -I 1-17>Iun 1&Jun 19Jun 20-Jun 21-Juri 22Wun 23-Jun 24-Jun 25^fun 2&Jun 27-Jun Date Figure A-3.11: Discharge (Q) and suspended sediment concentration (SSC) at Gauging Station C. 17-27 June, 1994. 216 Appendix A-4 : Discharge and Suspended Sediment Concentration at Russell Creek Gauging Station (WSC Stn. 08HF007) 100 -a) Discharge and SSC at Russell Creek, 30 September - 20 October, 1993 _ — — : — -1000 10 -; i -==, 1 ~ — -100 % E CJ CO Oi -10 1 E a i -' = z z—r__ — = •'• ~ i E = 0.1 - -: \ j i — i — i — <—i. —1 i i 1 | 30-S«jp 02-Oct 04-Oct 06-C i 1 i * r W O M W 10-Oct • 12-C Date 1 1 1 K* 1 4 0 d tS-C — i — i — i — t >ct irj-Od 20-C S S C Q (6-hr.) b) Discharge and SSC at Russell Creek, 20 October - 9 November, 1993 Figure A-4.1. Discharge (Q) and suspended sediment concentration (SSC) at Russell Creek Gauging Station, a) 30 September - 20 October, 1993; b) 20 October - 9 November, 1993. 217 E a) Discharge and SSC at Russell Creek, 9 - 29 November, 1993 - Runoff Event 6 - i 1 1 1 1-E. O M " " « • « • » 2 I N „ • 23-Nov 25-No, . 27-No. 2 W * > . D a l e S S C Q (est. daily) E. O b) Discharge and SSC at Russell Creek, 29 November - 19 December, 1993 Runoff Event 7 H 1 h n—4-29-No. 01-D«= CCH*c OWfcc 0 7 - & K I W > K 1 H > « 1 7 - 0 « 1S-0.5 Date S S C Q (6-hr.) Q ( e s t dai ly) Figure A-4.2. Discharge (Q) and suspended sediment concentration (SSC) at Russell Creek Gauging Station, a) 9 - 29 November, 1993;. b) 29 November - 19 December, 1993. 218 c a) Discharge and SSC at Russell Creek, 19 December, 1993 - 8 January, 1994 E, O 0.1 H I -1 1 1 — ^ 1 1 1 H *C»C 2 1 - 0 « : 2J-C.C 2 W * c 27.[*c 2 W * c 31 .C*: 0 2 J . n 0 * j T , r * J „ C * J « I Date • SSC Q (6-hr.) a b) Discharge and SSC at Russell Creek, 8 - 28 January, 1994 • 1 j i i — , , —; j Runo n mr vuirun t T c i n i / \ i / \ i \ \ _ ^ — j y r >v. ^=z===^ — i i i i | | i 1 i i , i • i 1 I > I 1 1 1 1 1 1 • — 1 — - f -E 06-Jan 10-J«n 12>l«n 1+Jart 16-Ji/i 18-J«n Date 20-Jan 22-Jin 24-J»n 2&Jin 26-Jan • SSC Q (6-hr.) Figure A-4.3. Discharge (Q) and suspended sediment concentration (SSC) at Russell Creek Gauging Station, a) 19 December, 1993 - 8 January, 1994; b) 8 - 28 January, 1994. 219 c a) Discharge and SSC at Russell Creek, 28 January - 17 February, 1994 t-f«b 13-F«tp 15^«b 17-F«t> SSC Q (6-hr.) b) Discharge and SSC at Russell Creek, 17 February - 9 March, 1994 • SSC Q (6-hr.) c Figure A-4.4. Discharge (Q) and suspended sediment concentration (SSC) at Russell Creek Gauging Station, a) 28 January - 17 February, 1994; b) 17 February - 9 March, 1994. 220 a) Discharge at Russell Creek, 9-29 March, 1994 100 - i • : 09-Mar 11.Mir 13-Mar 15-Mar 17-Mar 19-Mar 21-Mar Date 23-Mar 25-Mar 27-Mar 29-Mar b) Discharge at Russell Creek, 29 March - 18 April, 1994 100 "I r - i — — — 29-Maf 31-Mar 02-Apf 10-Apr 12-Apr 14-Apr 16-Apr 18-Apr Figure A-4.5. Discharge (Q) at Russell Creek Gauging Station, a) 9 - 29 March, 1994; b) 29 March - 18 April, 1994. No SSC data. 100 a) Discharge at Russell Creek, 18 April - 8 May, 1994 e a Runoff Event 16 • /\':;.:r:':: •+- •+- 4-18-A(» 20-Apr . 22-Ap» 26JM* •ate 30-»<X 02-M.y 04-May 06 -Mi , 08-Uiy b) Discharge at Russell Creek, 8-28 May, 1994 E. • > i • i • -i • i i I 1 1 1 1 1 j — 08-Miy 10-M«y 12-May 14-May 1(vMay 18-May 20-May 22May 2<-M.y Date Figure A-4.6. Discharge (Q) at Russell Creek Gauging Station, a) 18 April - 8 May, 1994; b) 8 - 28 May, 1994. No SSC data. 222 a) Discharge at Russell Creek, 28 May - 17 June, 1994 100 ::::::::::j:::::;;:;::t::::::':'i —-p—^^ 10 -; : ] r Runoff Event 17 £ O 1 -0.1 - 1 i ' L , ! . ; . ; ! 2 8 . M , , 3 0 - M „ 01-Jun O^Jun 05-Jun 07-Jun 0!>in H J u n , 3 . J u n i s j l i n ' ' Date b) Discharge at Russell Creek, 100 17-27 June, 1994 10 -£ O • 1 -0.1 -! 1 ' 1 i ' i 1 j , i , I , i . ; . - - ; 1 17-Jun ,9-jun 21-Jun 2 3 J u „ 2 ^ 2 7 . J l m ^ ^ ^ ' ' ' Date Figure A-4.7. Discharge (Q) at Russell Creek Gauging Station, a) 28 May - 17 June, 1994; b) 17 - 27 June, 1994. No SSC data. 224 JS u 0 B 1 o es u ~ c o CS .5 <u OA - S 3 CS o • cs H H = C - (ollsel) by lime period 30 Sep. -18 Nov. /93: C - H range: 2 mm H = C - 0.273 m 18 Nov.-20 Dec./93: C - H range: 9 mm H = C - 0.270 m 20 Dec./93-12Feb./94: C - H range: 10 mm H = C - 0.260 m 12 Feb.-xx Apr./94: C - H range: 10 mm H = C - 0.273 m 10 May-28 Jun./94: C - H range: 11 mm H = C - 0.286 m I — • E o — CM CM CM TJ- t?-r-- r- r~ r~ Is* CM CM CM CM CM • • ' O O O O O 0.274 0.274 0.265 0.265 cn cn cn co cn co CM CM CM O O O CO CO CO O O i to <C> r- r— CM CN CN CN O o o o d o co o •»— to O CO co co co at co o eo CM CM CM CM CM CM CN d ci ci ci ci ci ci i ? CO CM CO T CO CM Ol CO CM CM > -6 O O O O Q 0.112 0.111 0.158 0.143 0.143 0.226 0.075 >o <o cn «o N N tf O O O O O i - o O d O O O CO T cO CM CM CO CO to tO tO tO tO XT O p o o o o o d d d d d d d m 1 o »- co T - r-CO O O l O p~ , a ip io ui <n m ( o d o o o 0.574 0.634 0.616 0.722 O) lO lO IN. CM CJ> CO * -d d d o' N T - T - O W O l CO O O O * - o> o> tO lO to U " > tO -cr d d d o d d d o | * C O O N to O O l Ol CO CO , t . -CT CO CO CO CO o o o o o 0.386 0.385 0.423 0.408 0.408 0.481 0.340 CO TT r— TT LO •<r n- CN r» co co co ro ro O O d d O d CN TJ- CO TJ- r - CN r -tO CO CO T T T f CO co to to CO CO CO CO ci ci ci ci ci ci ci CO CD o ui e d O C M C M C M T - * - » - C O < 6 i b i o'ioir)V)vj^r O O CO CM 0> cb cb wi cri -cr* O CM Ol co ui v O O T O CO cb <6 to to ^ OJ OJ CQ (D K is. © IO IO i f l io ui V) V Time (PST) o o m co o o o cn TI- »— V cb o co cb co *— CJ .— .— CJ .— 13:18 14:05 12:00 10:35 10:50 16:00 11:37 O O i c O ui to y O TT "«r O CM CO CM tO CO T - ^ - T - T- O lO to to O O to o r- O tO IO CO ••— O . CO S T-' uS (D »- O D O O T- o - t— i — o Date (1993-94) 3 0 Sep. 16 Oct. 23 Oct. 25 Oct. 05 Nov. 06 Nov. 18 Nov. 18 Nov. 05 Dec. 07 Dec. 20 Dec. 20 Dec. 16 Jan. 12 Feb. 12 Feb. 27 Feb. 19 Mar. 10 May 10 May 12 May 15 May 16 May 28 Jun. c o u CU > C o u TJ es <U si si u » o c i, *3 •fe o <u OX) es CQ e o es OD C "So 3 es o cs H H = Adj. C - (offset) by time period 01 - 23 Oct. /93: C - H range: 0 mm H = C-0.145m 23 Oct. - 07 Dec. /93: C - H range: 23 mm H = C-0.149m 07 Dec./93-16 Jan./94: C - H range: 3 mm H = C-0.138 m 16 Jan. - 26 Feb. /94: C - H range: 0 mm H = C-0.139 m 26 Feb. -11 May /94: C - H range: 4 mm H = C- 0.147 m 11 May-27 Jun./94: C - H range: 23 mm H = C- 0.169 m I o f •0* < 0.145 o CM co CD -M- CO o ' d o o 0.136 0.139 0.139 0.148 0.145 0.149 O o r-co eo to cn o o o o X E (<0) 0.075 cn T f co co r~ r~ r- r~ o o o o o o o o 0.073 0.086 0.086 0.039 0.062 0.029 Ol CO CO co CM CM CM CO O O O O Ci O O O co E 0.389 oi co N r-CO CO CO CO CO CO CO CO O O O O 0.387 0.401 0.401 0.361 0.385 0.349 0.350 0.358 O < 0.030 0.220 m co co o co -~ T -CM CM CN CN O O o ci" 0.209 0.225 0.225 r- r— co co o r-CN -c-o o o cn co <n rt o O CO o> CN CN. T - -y ci ci ci ci o f 0.030 0.213 CO O O TT CM T- T- O CM CM CM CM O O O O 0.203 0.218 0.218 0.182 0.201 0.174 CO CM -"J- CO O O CO CO CN CM y— •»— d d d o C.S. Bait. (V) O l •»!• in cn o ^ q co* cn cn cn ***** °. cd cn* Ol CO* TT* oi *- -cr cri cri TT o> O l CO CO tO uS to T T Time (PST) 16:25 10:42 CM O O O TT O T f x -O CO* CO O 10:15 10:30 10:35 12:40 13:45 12:20 09:00 O to O tO O CM CO t o cn <J> CN co O O i - o Date (1993-94) 01 Oct. 23 Oct. 23 Oct. 05 Nov. 07 Dec. 07 Dec. 16 Jan. 16 Jan. 26 Feb. 26 Feb. 20 Mar. 11 May 1 11 May 15 May 27 Jun. I l l ro co o » 2 *S - £ <° •=r CD o 5 "3 ro §^ro C ™ C D ; i W O x V) O *E S3 2 2. tn ro « ro cn -o • (p) o • ro TJ Vi « • ) co -»-c 2. j CJ cn ? i c I <u — 3 ^ T J i ^ 2L i O O vi ro Q Si O 0 CJ (D « a: "° 225 Appendix B-2 : Stage-discharge rating formulae Table B-2.1. Rating curve data, Gauging Station B Date Q Meas. Staff-Gauge Notch Head Measured (1993-94) Method Stage (cm) (cm) Q (l/s) 25 Oct. MM 60.1 13.2 10.4 05 Nov. MM 59.3 12.6 10.4 18 Nov. B 57.4 11.1 5.7 05 Dec. MM 63.4 15.8 16.4 16 Jan. MM 72.2 22.6 55.1 12 Feb. B 52.9 7.6 2.1 19 Mar. B 56.5 10.4 5.6 Table B-2.2. Rating curve data, Gauging Station C Date Q Meas. Staff-Gauge Notch Head Measured (1993-94) Method Stage (cm) (cm) Q(l/s) 05 Nov. B • 38.7 ' " ' 7.3 2.3 07 Dec. B 38.7 7.3 2.1 16 Jan. B 40.1 8.6 3.7 20 Mar. B 38.5 6.2 1.4 11 May B 34.9 2.8 0.15 cn-203.wb1 Notes: 1. Q measurement methods: B = bucket retention, MM = Marsh McBimey current meter MM = Marsh McBimey current meter gauging. Formulae for conversion of staff gauge stages (S) to notch heads (H): (from Appendix B-1) Gauging Station B: H = 0.78 S - 0.337 m Gauging Station C: H = 0.95 S - 0.295 cm; before 26 Feb. /94 H = 0.95 S - 0.304 cm; after 26 Feb./94 Standard weir rating formulae (Church and Kellerhals 1970): 90-degree V-notch: Q = 1.4*HA2.5 (units m, s) Rectangular Q = C * W * H A 1.5 (units m, s) W = weir width -C = coefficient which depends on ht. of weir relative to depth of flow. Standard rating formula used where good fit with stream gauging measurements. Adjusted formulae used at higher flows where shooting flow - due to inadequate weir-pool size — resulted in rating formula underestimating flow measured by stream gauging. When flows exceeded capacity of V-notches, rectangular weir formula used with full-V flow added. Summary of formulae used, by gauging station: Gauging Station B: Q = 1400 * H A 2.5 (l/s) when H < 0.097 m Regular formula Q = 4500 * H A 3.0 (l/s) when 0.097 m <= H < 0.400 m Shooting flow, less than full-V Q = (3458 * (H - 0.400) A 1.5) + 288 (l/s) when 0.400 m <= H < 0.690 m Rectangular + full-V flow When H > or = 0.400 m, flow overtops V-notch, then add V-notch and rectangular components (see site diagram). V-notch component: Q = 4500 * H A 3.0 (l/s) Q = 4500 * 0.400 A 3.0 (l/s) Q = 288 (l/s) Rectangular component: Q = C * W * H A 1 . 5 W= 1.82 m C depends on weir ht relative to head (see Linsley and Franzini 1979): Weir ht. (pool bed to base of rectangular section of weir crest) ~ 0.6 m, Weir ht. (m) Head (m) C 0.60 0.06 1.85 0.60 0.12 1.88 0.60 0.30 1.91 Use.C = 1.9 (for m,s units), so C = 1900 (for l/s) Q = 1900*1.82*HA1.5 Q = 3458 * H A 1.5 (l/s) Combined components: Q =. 3458 * (H - 0.400) A 1.5 + 288 l/s When H > or = 0.690 m, flow overtops side wingwalls. Max. H, 30 Sep. /93 - 27 Jun. /94 : 0.513 m - OK. Gauging Station C: Q = 1400 * H A 2.5 (l/s) when H < 0.059 m Regular formula Q = 5000 * H A 2.95 (l/s) when 0.059 m <= H < 0.270 m Shooting flow, less than full-V Q = (1463 * ( H - 0.270) A 1 5) + 105 (l/s) when 0.270 m <= H < 0.530 m Rectangular + full-V flow When H > or = 0.270 m, flow overtops V-notch, then add V-notch and rectangular components (see site diagram). V-notch component: Q = 5000 * H A 2.95 (Us) Q = 5000 * 0.270 A 2.95 (l/s) Q = 105 (l/s) Rectangular component: Q = C * W * H A 1 . 5 W = 0.77 m C depends on weir ht relative to head (see Linsley and Franzini 1979): Weir ht. (pool bed to base of rectangular section of weir crest) ~ 0.4 m. Weir ht. (m) Head (m) C 0.40 0.40 0.06 0.12 1.89 1.91 Use C = 1.9 (for m,s units), so C = 1900 (for l/s) Q = 1900* 0.77*H A 1.5 Q = 1463* H A 1.5 (l/s) Combined components: Q = 1463 * (H - 0.270) A 1.5 + 105 l/s . When H > or = 0.530 m, flow overtops side wingwalls. Max. H, 01 Oct. /93 - 27 Jun. /94 : 0.279 m ~ OK. 228 Appendix B-3 : Turbidity - SSC Relations Table B-3.1. SSC Sample Data, Gully Gauging Stations Sample Info. Filter Filter Net Sample OBS Site Date Time Tare Total Mass Volume SSC Turbidity (93/94) (PST) (mg) (mg) (mg) (ml) (mg/I) (V) (NTU) Stn. B 25 Oct. 12:00 93.35 93.90 0.55 780.7 0.7 0.034 13.6 " • 26 Oct. 04:00 92.90 93.31 0.41 777.7 0.5 0.034 13.6 27 Oct. 20:00 93.13 93.78 0.65 764.1 0.9 0.032 12.8 28 Oct. 12:00 92.19 93.17 0.98 778.2 1.3 0.040 16.0 ** 29 Oct. 12:00 92.48 93.19 0.71 774.5 0.9 0.035 14.0 30 Oct. 12:00 92.56 93.20 0.64 769.0 0.8 0.030 12.0 ** 31 Oct. 12:00 92.79 93.67 0.88 778.2 1-1 0.035 14.0 ** 01 Nov. 12:00 93.07 93.77 0.70 770.0 0.9 0.032 12.8 ** 02 Nov. 04:00 91.50 92.48 0.98 795.4 1.2 0.031 12.4 05 Nov. 12:00 91.00 91.05 0.05 773.7 0.1 0.034 13.6 05 Nov. 16:00 93.40 93.68 0.28 787.5 0.4 0.033 13.2 05 Nov. 20:00 92.74 93.20 . 0.46 781.1 0.6 0.032 12.8 06 Nov. 00:00 92.16 92.63 0.47 782.4 . 0.6 0.032 12.8 06 Nov. 04:00 91.53 92.00 0.47 781.1 0.6 0.031 12.4 06 Nov. 08:00 92.40 92:79 0.39 779.6 0.5 0.031 12.4 ** 07 Nov. 12:00 90.97 91.31 0.34 777.1 0.4 0.031 12.4 15 Nov. 00:00 92.62 95.88 3.26 780.2 4.2 0.035 14.0 18 Nov. 00:00 92.15 92.67 0.52 782.4 07 0.035 14.0 22 Nov. 12:00 92.92 93.35 0.43 646.1 0.7 0.035 14.0 10 Dec. 00:00 92.63 95.22 2.59 .774.5 3.3 0.067 26.8 ** 15 Dec. 00:00 92.47 92.47 0.00 779.6 0.0 0.036 14.4 M 19 Dec. 12:00 92.52 92.48 0.00 779.0 0.0 0.041 16.4 ** 02 Jan. 12:00 206.15 211.24 5.09 792.5 6.4 0.033 13.2 ** 04 Jan. 12:00 206.70 207.65 0.95 793.5 1.2 0.036 14.4 ** 08 Jan. 12:00 206.25 210.38 4.13 787.0 . 5.2 0.036 14.4 23 Jan. 00:00 206.25 209.09 2.84 794.5 3.6 0.036 14.4 28 Feb. 18:00 206.25 216.74 10.49 780.0 13.4 0.047 18.8 01 Mar. 18:00 206.25 213.98 7.73 785.0 9.8 0.049 19.6 02 Mar. 18:00 206.25 214.40 8.15 781.0 10.4 0.070 28.0 27Mar. 18:00 206.25 214.95 8.70 789.5 11.0 0.045 18.0 28 Mar. 18:00 206.25 213.51 7.26 798.5 9.1 0.051 20.4 11 Apr. 18:00 206.25 216.38 10.13 775.5 13.1 0.067 26.8 Stn. C 23 Oct. 11:00 92.02 97.00 4.98 496.0 10.0 0.039 15.6 05 Nov. 16:00 91.80 92.71 0.91 503.9 1.8 0.040 16.0 07 Dec. 10:10 91.77 91.77 0.00 510.4 0.0 0.038 15.2 16 Jan. 11:15 206.18 206.84 0.66 511.5 1.3 0.039 15.6 cn-212.wb1 Notes: 1. Filters used in lab analysis: 1993 samples, 1.2 microns; 1994 samples, 2.7 microns. 2. OBS turbidity readings are 15-min. averages 3. Sample collection methods: Stn. B samples, Isco pump sampler; Stn. C samples, grab samples. Table B-3.2. SSC Sample Data, Russell Creek Gauging Station Sample Info. SSC 15-min. OBS Sample Info. SSC 15-min. OBS Date Time (mg/I) Turbidity Date Time (mg/I) Turbidity (93/94) (PST) (NTU) (93/94) (PST) (NTU) 22 Oct. /93 09:16 4 15.7 15 Jan./94 02:05 1 43.1 27 Oct. /93 08:38 4 86.3 28 Feb. /94 09:22 179 39.2 27 Oct. /93 08:53 1 266.7 28 Feb. /94 09:47 51 354.9 27 Oct. /93 09:08 1 78.4 28 Feb. /94 10:02 51 425.5 27 Oct. /93 09:23 2 78.4 28 Feb. /94 10:17 44 452.9 27 Oct. /93 09:25 2 78.4 28 Feb. /94 10:32 42 119.6 27 Oct. /93 09:26 2 78.4 28 Feb. /94 10:47 45 337.3 27 Oct. /93 09:28 89 78.4 28 Feb. /94 11:02 47 88.2 27 Oct. /93 09:29 8 78.4 28 Feb. /94 11:17 43 270.6 27 Oct. /93 09:31 4 213.7 28 Feb. /94 11:32 40 358.8 27 Oct. /93 09:32 2 213.7 28 Feb. /94 11:47 37 37.3 27 Oct. /93 09:34 2 213.7 28 Feb. /94 12:02 33 407.8 27 Oct. /93 09:35 2 213.7 28 Feb. /94 12:17 33 35.3 27 Oct. /93 09:37 1 213.7 28 Feb. /94 12:32 27 307.8 27 Oct. /93 09:38 2 213.7 28 Feb. /94 12:47 25 35.3 27 Oct. /93 09:40 1 213.7 28 Feb. /94 12:49 25 35.3 27 Oct. /93 09:41 1 213.7 28 Feb. /94 12:50 25 35.3 27 Oct. /93 09:43 1 213.7 28 Feb. /94 12:52 26 35.3 27 Oct. /93 09:45 1 213.7 28 Feb. /94 12:53 24 35.3 27 Oct. /93 09:46 0 62.7 28 Feb. /94 13:01 24 70.6 27 Oct. /93 . 09:48 0 62.7 28 Feb. /94 13:03 20 70.6 27 Oct. /93 09:49 2 62.7 28 Feb. /94 13:14 22 70.6 27 Oct. /93 09:51 0 62.7 28 Feb. /94 13:15 19 70.6 27 Oct. /93 09:52 2 62.7 28Feb./94 13:16 21 51.0 01 Dec. /93 12:38 75 23.5 01 Mar. /94 11:10 11 29.4 03 Dec. /93 03:08 88 62.7 02 Mar. /94 05:30 81 41.2 03 Dec. /93 03:38 87 113.7 02 Mar. /94 06:00 111 311.8 03 Dec. /93 04:08 137 405.9 02 Mar. /94 06:30 113 427.5 03 Dec. /93 04:38 184 227.5 02 Mar. /94 06:32 108 460.8 03 Dec. /93 05:08 219 282.4 02 Mar. /94 06:39 121 460.8 03 Dec. /93 05:53 333 500.0 02 Mar. /94 06:47 119 49.0 03 Dec. /93 06:23 299 21.6 26 May /94 11:50 6 27.5 03 Dec. /93 07:08 253 ..• 78.4 27 May /94 06:26 2 56.9 03 Dec. /93 07:33 219 360.8 27 May /94 06:56 1 56.9 03 Dec. /93 07:34 212 360.8 27 May /94 07:41 3 41.2 16 Dec. /93 15:23 11 23!5 27 May /94 08:11 2 49.0 30 Dec. /93 01:35 1 25.5 30 May /94 06:41 1 31.4 30 Dec. /93 02:05 0 105.9 01 Jun. /94 06:26 1 39.2 01 Jun. /94 07:11 1 11.8 cn-501.wb1 Notes: 1. SSC sample data obtained from Sediment Survey of Canada laboratory. 2. All SSC samples collected by Isco pump sampler. 3. Turbidity data extracted from raw WSC data dump files. Appendix C - l : Gully Morphology Surveys Table C- l . ' l . Gully B Morphology Survey LB RB Gully Dimensions xs Dist. Slope Slope Run Rise Slope Slope Run Rise Depth Width Area # (m) dist. (m) (deg.) (m) (m) dist. (m) (deg.) (m) (m) (m) (m) (m2) Main C iully Stn. B 0 — _ _ _ 1 33 4.6 13.0 4.5 1.0 3.8 18.2 3.6 1.2 1.1 8.1 556 2 105 3.2 19.2 3.0 1.1 3.1 17.3 3.0 0.9 1.0 6.0 391 3 164 4.0 19.7 3.8 1.3 10.5 30.2 9.1 5.3 3.3 12.8 873 TS120 202 - - - - — — _ . _ 4 236 8.8 16.8 8.4 2.5 10.3 31.7 8.8 5.4 4.0 17.2 1001 5 285 5.6 21.3 5.2 2.0 5.6 35.0 4.6 3.2 2.6 9.8 598 6 358 6.7 -1.5 6.7 -0.2 4.3 6.7 4.3 0.5 0.2 11.0 669 7 407 5.6 18.3 5.3 1.8 6.9 23.3 6.3 2.7 2.2 11.7 685 TS120C 441 - - - - — — _ „ 8 472 18.0 43.0 13.2 12.3 18.9 35.8 15.3 11.1 11.7 28.5 2006 9 551 From survey map 26.3 25.8 From survey map 30.4 21.3 23.5 56.7 4742 10 639 From survey map 31.0 24.9 From survey map 26.6 28.1 26.5 57.7 4959 11 723 From survey map 23.4 34.2 From survey map 56.5 38.8 36.5 79.9 4112 12 742 Geodiri neter 34.5 32.7 Geodir neter 42.7 32.5 32.6 77.2 5660 13 870 5.9 42.5 4.3 4.0 10.0 42.0 7.4 6.7 5.3 11.8 1019 14 915 7.4 22.7 6.8 2.9 12.9 23.3 11.8 5.1 4.0 18.7 1302 TS53 932 - - - - — _ • — _ Source 962 - - - - - - -- - - - -Trib. G Lilly Confl. 472 — _ _ _ Trib. 1 529 18.3 45.8 12.8 13.1 16.4 45.5 11.5 11.7 12,4 24.3 2611 Trib. 2 630 7.0 45.5 4.9 5^ 0 18.4 37.2 14.7 11.1 8.1 19.6 1293 Trib. 3 661 4.9 28.8 4.3 2.4 3.9 46.8 2.7 2.8 2.6 7.0 431 TS120D 708 — — — — - - - - - -cn-3O0 wb1 Total area within gully crests (ha): 3.3 231 Table C-1.2. Gully C Morphology Survey LB RB Gully Dimensions xs Dist. Slope Slope Run Rise Slope Slope Run Rise Depth Width Area n (m) dist. (m) (deg.) (m) (m) dist. (m) (deg.) (m) (m) (m) (m) (m2) Stn. C 0 1 7 14 5 23.5 13.3 5.8 8.4 30.3 7.3 4.2 5.0 20.5 491 2 41 5.2 23.7 4.8 2.1 3.0 45.0 2.1 2.1 2.1 6.9 354 3 110 4.8 23.3 4.4 1.9 7.9 21.7 7.3 2.9 2.4 11.7 966 4 205 6.9 16.5 6.6 2.0 5.3 27.5 4.7 2.4 2.2 11.3 822 5 255 6.0 24.0 5.5 2.4 7.5 23.3 6.9 3.0 2.7 12.4 536 6 292 5.6 16.7 5.4 1.6 4.2 22.3 3.9 1.6 1.6 9.2 467 7-Source 356 6.0 11.2 5.9 1.2 5.4 13.3 5.3 1.2 1.2 11.1 355 cn-300.wt>1 Total area within gully crests (ha): 0.4 232 cn cu u u 3 o CO •*-> c cu s •5 CO T3 ' cu "S. S CO « c o 3 CU _N C/5 1 eu "3 u « Pk 1 u •5 e cu a a < eu CO cu CJ I* 3 © CO c cu S •3 cu CO T3 cu "c. E « "5 M U cu IB a H c o o CO CD a cu CO 03 Q . •E ta CO E E m a. o c o a> TO E cu 0 CO d) c 1 ro £1 <L cu ro m 0: E 0 E (M J _f O © CU O TO t 3 CO "E o o CM CO 00 T3 CU £ 1 cu > o TO E o o 10 O T5 © co" ro o CO CO o O o CN CO CO CQ E ro 0 3 cu > o TO E ro Q O _i O ) ba o a> O . o IN CO H > o . Q ro cu o ro a> "E TO 1— O ) C O c CU CO TO O C L •g E a> a. o CO CO _2 ro H -4-CQ ro E TD 03 co o CL CU •o lo CU c T •* E cu e o N E ro - 3 Q O _i ca •o c !c cu £ } CU o> •a cu 5 c o 'co o CL cu Q 10 ca ro cu c cu o TO ro •o cu 1 O cu Dl "cu X3 CU c o N •tt o a. CO c ro ro o CO cu _3 .TO TO O CO - <u 2 TO O CO .2 2- = CU TO S CO W — TOCO JQ — cu Q . o CO CO _2 TO I— CD CQ TJ CU c O to 03 ro CQ ro CU •g 'co tz o o CO 0 0 CQ 03 ro 03 fc? ro o o cu CL CL 3 03 w " 3 E ib c ro 3 y 03. T3 => O >> CM 3 O o 03 >» _ro cu e 03 o 03 w 3 E 10 c ro 3 o 03 T3 C ID CO O 233 E E CM ro « -a cu « cu c 3 u H CO C .2 -«—» 3 JO CU • U cu « 1 Total O OO 00 00 O O) 00 o o o o i o> o i 6 oi d o o 0 0 ) 0 1 0 ) 0 0 ) 0 ) o o Y— . T— T— T— jre class) 1 Clay co o o i s q n m co r -CN T-" Tf" CO CO tN O oi rVentwbrth texti 1 Silt O 'T S O) CO CO . . . . co" CO CO CO CO LO >^S5 CO CO o o + + "co, 'to s (% mass by V I Fine sand (N CO O CD OO N C O O UTi ^ ^ co in oo T-" o* CN ;ize distribution I Med. sand CO CO CO CO O) O CO f - CO N ^ U ) O ) CO CN CN T- CN Particle s | Coarse sand T f r-. r-- r-~ T f co o> CD o co" cd CN to' CN o' ai •«-* T- t - M i - i - r CO Pebbles O CN CO CO T— T— 00 CO CO LO O O) oi oi Tf" LO LO o N N CO CM CO CO CD CO T-Sample T-COTJ-IOCOO-OO CM CO m m m m m c Q c Q o o E E fN « •a cu CI CJ s 3 H CO S O •*-» 3 CU _N CO CO CM U cu 23 CJ H Organic content (%) CO CN CN T- TJ- CNI LO O O CO O CO CO* T - r — co" to CO T- CN CN 1 Total sed. oo co co r>- co oo LO O O co o> co co oi oi oi O O CO O) CO CO CO O) CO o o T~ T— ;ure class) 1 Clay <1 CO T f N ( v . (N CO O-T-T— CO* CO CD" oi oi f*~* T~ o Wentworth texl 1 Silt T- LO oq LO r>- CN LO C N T ^ O O O ' C N C O ' I O >; >; T— T- T- T- T- r- r - CO CO o o + + ; (% mass by I Fine sand OO O r - CO CO 00 LO ( N T-T-" T f " CO T - T f T f IO T-" o' CN T— T- T— •«- T- CN T- CO ize distributions I Med. sand O LO CO 00 CO CM CO T f T-CO T f LO co* oi LO" oi 00 LO CN T- T- T- T- CN T- T— CN Particle si I Coarse sand co T— CN oo T f T f oo r--. co i r i c o c N c o ' c o ' T f T - " cd rf CN LO T f LO T f CO CO CD CO Total sand -<f eq CN LO CD T f r>- co o> cd T f rf oi s T f co cd oi r> co N co N r> r> o> oo Sample T - « T f m to N co CM co CO CQ CQ CQ CQ CQ CQ O O - «' 234 Appendix C-3 : Gully Gauging Station Site Sketches FLOW Faceplate SCALE 1 : 25 All dimensions in cm. b) Elevation View Pool Bed . Figure C-3.1. Gauging Station B site sketch, a) Plan view, b) Elevation view, looking down-stream from weir pool. 235 -44-FLOW Faceplate "77 53 V-notch Instrument Housing Weir Pool v O B S T <N a) Plan View SCALE 1 : 10 All dimensions in cm. b) Elevation View 00 CM Pool Bed Figure C-3.2. Gauging Station C site sketch, a) Plan view, b) Elevation view, looking down-stream from weir pool. 236 

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:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0087011/manifest

Comment

Related Items