UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The environmental history and geomorphic impact of 19th century placer mining along Fraser River, British… Nelson, Andrew David 2011

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

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
24-ubc_2011_spring_nelson_andrew.pdf [ 29.01MB ]
Metadata
JSON: 24-1.0071772.json
JSON-LD: 24-1.0071772-ld.json
RDF/XML (Pretty): 24-1.0071772-rdf.xml
RDF/JSON: 24-1.0071772-rdf.json
Turtle: 24-1.0071772-turtle.txt
N-Triples: 24-1.0071772-rdf-ntriples.txt
Original Record: 24-1.0071772-source.json
Full Text
24-1.0071772-fulltext.txt
Citation
24-1.0071772.ris

Full Text

The Environmental History and Geomorphic Impact of 19th Century Placer Mining along Fraser River, British Columbia by Andrew David Nelson B.A. Geology, The Colorado College, 2007 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in The Faculty of Graduate Studies (Geography) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) April 2011 ©  Andrew David Nelson, 2011 ii ABSTRACT One possible source of part of the 171,000 to 229,000 m3 of gravel that accumulate annually in the gravel bed reach of Fraser River in the Lower Mainland is sediment that was dumped into the river by 19th century placer gold mining.  Historical data suggest that, following the Fraser Gold Rush of 1858 and the Cariboo Gold Rush of the 1860s, a substantial placer gold extraction industry was established and continued into the beginning of the 20th century.  Gold production figures and typical gold concentrations can be combined as a proxy to estimat that around 50 million m3 of sediment were excavated by mining activity. Excavations caused by mining are still present in the modern landscape.  The areas covered by 456 mine excavations were mapped between Hope and the Cottonwood Canyon along Fraser River.  A subset of 58 mines was surveyed and strong regression relations predicting mine volume from mine area were found and used to produce estimates of the volume of excavated material.  This allows estimation of the total excavated volume of sediment: 45,900,000 m3.  Small mines (<315,000 m3) contributed most of the tailings; and only 30% of the tailings came from hydraulic mining.  Grainsize sampling and stratigraphic observations suggest average mine tailings were composed of 14% small cobbles, 32% gravel, 41% granules and sand, and 13% silt and clay.  The resulting sediment wave on the Fraser can be classified as a megaslug.  Sediment transport calculations suggest that the capacity of the Fraser to transport sediment is substantially higher than the average tailings load, so the key factor limiting downstream movement of sediment and resulting delivery to the aggrading reach is the virtual velocity of the sediment. Annual velocities of between 1 and 5 km a-1 are probable. These velocities predict 100,000 to 700,00 m3a-1 tailings are delivered to Hope, which compares favorably with the observed aggradation rate.  Sediment from placer mining on the main stem of the Fraser may continue to influence the rate of sediment delivery to Hope for another century or more, nevertheless, the historical aggradation rate may not represent future conditions. iii PREFACE Parts of this thesis were facilitated by collaboration.  Map 2.1 and the research supporting it were a collaborative effort of the author of this thesis, Mike Kennedy, Eric Leinberger, and Robert Galois. Contributions of  Mike Kennedy and Robert Galois to the supporting research are described in Chapter 2.  Geocontextualization of archival records was a collaborative effort of the author of this thesis and Mike Kennedy.  With a few exceptions, I placed historical references from High Bar upstream and Mike Kennedy placed historical references from Big Bar Creek downstream.  Eric Leinberger was the cartographic designer. iv TABLE OF CONTENTS ABSTRACT .......................................................................................... ii PREFACE ........................................................................................... iii TABLE OF CONTENTS     ..........................................................................   iv LIST  OF  TABLES ................................................................................  vii LIST OF FIGURES  ...............................................................................  viii LIST  OF  MAPS  ..................................................................................  x ACKNOWLEDGEMENTS ........................................................................  xi DEDICATION  .....................................................................................  xii CHAPTER 1: INTRODUCTION  ................................................................  1 Research Objectives  .........................................................................  2 Historical Environmental Reconstruction and the Use of Historical Data in Geomorphology ..............................................................................  3 Placer Mining Impacts on Geomorphic Systems  ................................  5 Methodology for Using Historical Sources in Environmental Science  .........   9 General Project Methodology .............................................................  14 CHAPTER 2: HISTORY OF PLACER MINING IN THE FRASER WATERSHED ...................................................................  18 Fraser River Gold Mining Landscape .....................................................  18 Initial Discovery and the Gold Rushes .....................................................  21 Development of the Mining Industry Along Fraser River .........................  25 Yale Vicinity: Maria Bar to the Hell’s Gate Canyon ..................................  25 Boston Bar Vicinity ......................................................................  27 Lytton Vicinity: Fraser River from Eight-Mile Hill to Van Winkle Flat and Thompson River ............................................................  27 Lillooet Vicinity: From north of Van Winkle Flat to Pavilion ........................  29 Grasslands: Pavilion to the Sheep Creek Bridge ...................................  31 Quesnel Forks and Alexandria: Sheep Creek Bridge to Cottonwood Canyon .................................................................  31 Quesnel River: Quesnel and Cariboo Rivers Below Quesnel and Cariboo Lakes  .................................................................   32 Quantitative Information Regarding the Geomorphic Impact of Placer Mining  .....  33 Conclusion  ......................................................................................   40 CHAPTER 3: MAPPING AND MEASURING MINES ...................................... 42 Methods for Mapping Physical Evidence of Mining  ....................................  42 vGeographic Variations in Methods  .......................................................  46 Summary of Physical Evidence of Mining  .................................................  48 Mine Surveys  ..................................................................................  50 Site Selection  ............................................................................  51 Survey Methods  .........................................................................  51 Surface Creation and Volume Estimation  ..........................................  52 Survey Results  ..........................................................................  54 Regression Relation Predicting Mine Voume From Mine Area  .....................  62 Conclusion  .....................................................................................  66 CHAPTER 4: ESTIMATING THE TOTAL CONTRIBUTION OF GRAVEL TO FRASER RIVER  ...............................................................................  67 Grain Size Distributions  .....................................................................  67 Methodology  .............................................................................  68 Results  .....................................................................................  70 Mine Site Stratigraphies and Facies Distributions  ...........................  70 Bulk Sampling  .......................................................................  71 Wolman Sampling  ..................................................................  73 Analysis  ....................................................................................   76 Estimating Grain Size Distributions for Each Facies  .......................  76 Constructing a Synthetic Grain Size Distribution to Represent All Material Dumped Into the River by Placer Mining  ...........................  77 The Total Contribution of Gravel to Fraser River  ........................................  78 Mine Volume Calculation Methodology  ..............................................  78 Results  .....................................................................................  79 Comparison With Historical Information  ............................................  79 CHAPTER 5: THE FATE OF EXCAVATED GRAVEL  ...................................  85 Physical Evidence  .............................................................................  86 Sediment Slug Behavior  ......................................................................  91 Introduction  ...............................................................................  91 Theoretical Background  .................................................................   91 Implications for Fraser River  .................................................................  94 Classification of the Fraser Slug  ........................................................  94 Sediment Transport Modeling to Estimate the Fraser and Quesnel Rivers’ Capacities  .........................................................  96 Virtual Velocity  ............................................................................  98 CHAPTER 8: CONCLUSION .................................................................. 102 vi WORKS CITED  ...............................................................................  105 APPENDIX A: MINE SITE SURVEYS  ................................................  116 APPENDIX B: GRAIN SIZE SAMPLING DATA  ....................................  128 APPENDIX C: LIST OF ALL MAPPED MINES  ........................................  131 vii LIST  OF  TABLES 1.1: Key placer gold regions in British Columbia  ....................................................  8 1.2: Internal checks of reliability  ...................................................................  11 2.1: Sources used in the creation of Map 2.1  .......................................................  20 2.2: Estimates of key economic and practical constraints on different methods of placer mining  ......  34 2.3a-b: Gold production values for the province of British Columbia and several regions along         and tributary to the Fraser River  ..................................................................  36 2.4: Estimates of gold production from Haggen (1923) and corresponding estimates of gravel            excavation .........................................................................................  38 2.5: Estimates of the proportion of gold extracted in different regions and time perods by rocker boxes, sluicing, and hydraulic mining  ............................................................ 38 3.1: Landscape characteristics indicative of historical mining sites  ................................. 43 3.2: Air photos consulted during mapping  ........................................................... 45 3.3a-d: List of mine site surveys  ...................................................................... 55 3.4: Average geometry of mines by technology  ..................................................... 58 3.5: Example 95% confidence and prediction intervals spanning the range of mine site areas  for both the sluice and hydraulic and groundsluice regressions.  .......................................... 65 4.1:  Percent by weight of grains in each facies that belong to various grain size classes  ................ 77 4.2: Various estimates of sediment excavation associated with placer mining.  ......................  82 5.1: Hydraulic conditions at various flood flows at Hope and Margeurite gauging stations.  ....... 89 5.2: Input grain size distributions for sediment transport calculations  ............................. 95 5.3: Calculated average annual bedload transport based on various input grain size  distributions      at three locations.  ................................................................................ 98 5.4 Lengths of important canyons along the Fraser below Quesnel  .................................... 99 viii LIST OF FIGURES 1.1: Typical sluice box (A), ground sluice (B), and hydraulicked (C) placer sites  ...................... 6 2.1: General locations of administrative boundaries, region names, and some of communities  ....... 19 2.2: Interrelation of proxies for placer mining gravel excavation volumes and practical and      economic constraints  .......................................................................... 34 2.3: Estimated total amount of gravel processed and the method of mining in several regions          through time  ..................................................................................... 39 3.1: Histograms of the logarithmically transformed variables area and volume for surveyed mine              sites  ............................................................................................... 59 3.2: Rendered views of some selected mine site surveys and reconstructed pre-mining survaces used         to estimate the depth of excavations  ............................................................. 60 3.3: Plot of regression relations predicting mine site volue from area for sluice (blue line) and        groundsluice and hydraulic (green line) geometries  ............................................ 63 3.4: Regression error residuals for relations predicting mine volume from mine area for hydraulic         and groundsluice (left panel) sluice (right panel) and geometries  ............................... 64 3.5: 95% confidence intervals and 95% prediction intervals  for regression relations predicting mine         volume from mine area  .......................................................................... 64 4.1: Example mine stratigraphies  .................................................................... 70 4.2: Box and whisker plots of the depth of each facies  ................................................. 71 4.3: Cumulative distribution plots of all bulk mine samples by facies  .................................. 72 4.4: Cumulative distribution plots of the fluvial gravel facies superimposed on plots of grain size       distributions from the lower Fraser, the Fraser at Lilloet, and Quesnel rivers  ................ 72 4.5: Cumulative distribution of bulk samples from mines of the debris flow/ colluvial gravel facies        superimposed on debris flow samples of Jordan (1994) and bulk fluvial gravel samples from            mines  ............................................................................................. 73 4.6: Cumulative distributions showing results of Wolman counts on mine lag deposits  .......... 74 4.7: Histogram of the maximum grain sizes observed for all sites where Wolman counts         were done  ........................................................................................ 74 4.8: Comparison of cumulative distributions of bulk samples from unsorted material and            Wolman counts from corresponding lag deposits  ................................................. 75 4.9: Comparison of the largest stone observed in Wolman counts of lag deposits with the          largest   stone in corresponding bulk samples.  ..................................................   75 4.10: Cumulative distribution plots of representative grain size distributions for each facies.     76 4.11: Longitudinal distribution of gravel inputs to the Fraser  .................................... 80 4.12: Proportional symbol map of estimated mine excavation volumes along Fraser River            between Hope and Cottonwood Canyon  .......................................................  81 ix 4.13: Frequency magnitude distribution of mine site volumes  ...................................... 83 5.1: Tailings dump of the Bullion Pit Mine into the Quesnel River  ............................. 85 5.2: Part of the large volume of tailings stored below the Hydraulic Mine located on the left       bank of Fraser River 11.5 km south of Quesnel  ............................................... 85 5.3: Cumulative grain size distribution plots of Wolman samples from an eroded and censored        tailings fan from below a mine (Van Winkle Flat) and from a newly formed bar below a        recent landslide, compared with plots of the grain size distributions mobilized by flood flow          conditions on the Fraser  .......................................................................... 87 4.4: Paired historical (1886) and modern (2009) photos looking downstream from the left bank         of the Fraser above Lillooet  ..................................................................... 88 5.5: Visual definitons of translation, dispersion and migrating inflection point  ................... 92 5.6: Partial duration series for flows on the Fraser and Quesnel rivers  ............................. 96 5.7 Cross sections at locations of sediment transport calculations. See text for      description of data sources  ...................................................................... 97 5.8: Estimated gravel delivery to Hope assuming unlimited capacity and virtual velocities       as specified in the figure  ........................................................................ 100 xLIST  OF  MAPS 2.1: Historical Minesites Along Fraser River from Hope to Quesnel Forks  .....................  142 xi ACKNOWLEDGEMENTS I offer my sincere gratitude to the faculty, staff and fellow students at the UBC who have deepened my understanding of the world.  I owe particular thanks to Dr. M. Church, who has patiently advised me through this research process and taught me the immense value of pragmatism.  Thanks also to Dr. M. Hassan and Dr. C.  Harris for their helpful insights into this research and writing. I thank Mike Kennedy, of Lillooet, for his incredible generosity in sharing his personal research, home, and love of the history and geography of mining along the Fraser; and the innumerable people who live and work along the Fraser who invited me onto their land and into their lives. This research was supported by Discovery Grant A7950 from the Natural Sciences and Engineering Research Council of Canada to M. Church and by the University of British Columbia Graduate Fellowship #6301 awarded to Andrew Nelson. Many thanks to my wife, Rachel, for correcting countless misspellings. xii To “the creature” who will hopefully live in a world just as wondrous and inspiring as the one I know. 1CHAPTER 1: INTRODUCTION  In the mid-late 19th century placer gold was extracted from along Fraser River and its tributaries.  Huge volumes of gravel were excavated from the banks and terraces along the river in pursuit of this gold, with consequences for Fraser River that this thesis seeks to investigate .  Placer mining in the Fraser drainage basin began in 1858 and continued into the beginning of the 20th century. Along the Fraser, miners worked surficial alluvial deposits, the locations of which have not generally  been determined (but see Kennedy 2009).  This thesis describes a fifty-year long mining industry along nearly six hundred kilometers of Fraser River.  Using two independent methods, it also estimates the volume of gravel that was excavated by mining.  First, it draws on historical information to produce an estimate based on gold production and on typical gold concentrations for the different types of mining. Second, it relies on an inventory of mine excavations along the river to directly quantify the volume of excavated gravel.  Two key conceptual issues are raised by the possibility of a large 19th century geomorphic disturbance associated with placer mining.  The first is the potential for a historical legacy that may be influencing current processes in the river.  Without historical context, modern process studies may reach erroneous conclusions regarding natural processes because the geophysical event sequences on which they are based are frequently much shorter than the timescale of trends in geomorphic systems (Church 1980).  Historical study such as that undertaken in this thesis has the potential to extend both qualitatively and semi-quantitatively the geophysical event sequences on which modern process studies are based.   James (2010) has observed the presence of legacy placer mining sediment in California and argues that this sediment is a major factor controlling current geomorphic processes. This thesis describes another location where historical placer mining sediment may substantially influence modern geomorphic processes.  The second issue raised by the consideration of historical placer waste sediment  along the Fraser is the possibility of spatial translation of the impact of the mining by downstream sediment transport.  Previous work considering the evolution and propagation of sediment slugs (eg. Nicolas et al. 1995) provides a basis for  evaluating possible connectivity between locations of placer mining and downstream reaches of the Fraser.  This work includes studies that consider the evolution of sediment slugs at relatively small and detailed scale (eg. Lisle et al. 2001) and other studies that consider general sediment virtual velocities (eg Beechie 2001) and movement of sediment through complex channel networks (eg Hooke 2003).  On the Fraser, it is possible that discrete sediment inputs over a very long stretch of river coalesced into a low amplitude wave which, upon entering the Lower Mainland, has caused pronounced aggradation. Because individual sediment inputs to the Fraser from placer mining are generally distributed and relatively small, the confirmation  of coalescence of discrete sediment inputs into a single slug would provide support for the hypothesis that placer waste sediment may be contributing to noticeable aggradation in the lower gravel-bed reach (Church et al. 2001). Lisle et al. (2001) speculate that waves may change behavior where the slope of the river decreases. “Aggradation,” they suggest, “may become pronounced downstream, 2for example, as a wave front advances into a reach where a decrease in channel gradient causes the ratio of storage to transport resulting from an increase in sediment load to increase” (p. 1415).  If the hypothesized phenomenon could be demonstrated, it would be another example of coalescence of discrete sediment sources into a single wave, and would be a unique test of the proposition that wave amplitude may increase where river slope decreases.  This work has the potential to make a significant contribution to studies of episodic sediment transfer (eg. Gilber 1917, James 2006, James 2010) and may help establish the applicability of the sediment slug paradigm at a very large scale. Fraser River drains an area of ~200,000 km2 and the hypothesized wave could have a length of 600 km. This is much larger than the longest wavelength studied hitherto (Knighton 1989).  This study faces similar problems to Knighton’s study where there was a “large number of widely-separated supply points and ... lengthy river distance (75 km) and time period (110 years)” (p.343 ).  It is possible that  a legacy effect from placer mining may be influencing modern processes along an aggrading reach of  Fraser River between the towns of Mission and Agassiz.  Spatial patterns in the aggradation indicate in general that the (upstream) area between Aggasiz and Laidlaw is degrading, while the area (downstream) between Aggasiz and Mission is aggrading (Church et al. 2001) and historical air photos show a zone of instability that has propagated downstream over the 20th century (Church and Ham 2004).  This suggests that there was an “influx of high volumes of bed material between the late 19th century and early 20th century, and a substantial reduction of incoming bed sediment since.”  Church and Ham (2004) observed the possibility that the recent sedimentation history in the gravel bed reach may be a legacy effect and suggested that the sedimentation history is associated with neoglaciation and “perhaps more significantly ... placer gold digging on the river bars and terraces and the major railway building projects along the Fraser and Thompson rivers.”  This thesis focuses on placer mining as a potential source of currently accumulating  sediment.  Other known anthropogenic sediment disturbances in the Fraser basin include road and railroad construction, forestry, minor urbanization, and agriculture.  The importance of understanding a possible legacy effect from placer waste sediment comes into particular focus where aggradation in the lower gravel-bed reach of the Fraser exacerbates a significant flood hazard.  Management decisions related to gravel extraction volumes for flood control are being made on the basis of historical information.  Gravel extraction must be carefully targeted so as not to overly disturb the river morphology and associated salmon habitat (Church et al. 2001, Church 2010). Forecasts based on the historical aggradation rate without understanding of the historical geomorphic context and the possibility of legacy sediment effects may lead to substantial damage to the river system.  Research Objectives  Because of the large spatial and temporal scales associated with the hypothesized sediment slug in Fraser River, the research described here will probably not contribute to the detailed theory of 3sediment waves.  Research of this slug will, rather, explore the applicability of the paradigm at a very large scale and study the dynamics of a wave under the particular boundary conditions associated with Fraser River.  However, understanding the dynamics of sediment associated with placer mining will have important implications for management of the aggrading reach of the river.  The most fundamental question of interest for this research is: has a sediment slug associated with 19th centry mining activity traversed Fraser River? The work of Church et al. (2001) and Church and Ham (2004) shows downstream propagation of a zone of instability, with degradation in the upper portion of the alluvial gravel-bed reach of the river, and aggradation in the lower portion of the alluvial gravel-bed reach.  These observations indicate that there may be a slug moving down the lower river today.  This research will explore antecedents that may have contributed to these 20th century phenomena.  The more focused question of interest for this research is: how much of the current excess gravel in the aggrading reach can be explained by placer mining? Studying three sub questions will contribute to answering this overarching question: 1) How much waste from placer mining entered the river? 2) What is the grain size distribution of that waste? and 3) (How) did the river move the sediment? Historical Environmental Reconstruction and the Use of Historical Data in Geomorphology  Attempting to understand the environmental impact of placer mining along Fraser River is a task of historical environmental reconstruction and historical geomorphology. Substantial documentary evidence exists of placer mining activity along the Fraser and its tributaries, and a body of literature discusses approaches to the use of historical and documentary data in geomorphic study. Therefore,  it is useful to provide a review of work in these fields in order to contextualize methods used in this project.  This review will consider  the importance of historical understanding in modern day process studies, human impacts on geomorphic systems, and specific examples of the impact of sedimentation associated with mining, and will conclude by outlining a general methodology for pursuing historical information regarding physical systems and human alterations of those systems.  Allan James (1999), who has worked extensively studying the history of placer mining and its persistent impacts in the Sierra Nevada, argues that a historical consciousness is critical as a foundation for both modern process studies and river engineering attempts.  James states that in the Sacramento River system, “in spite of extensive geomorphic changes, a rich body of historical evidence, and extensive engineering research on the hydraulics of the system, there has been little understanding of the importance of historical alluvium to on-going channel changes by planners and engineers. Vast deposits of 19th century mining sediment continue to influence flooding and sediment yields in the mountains and Sacramento Valley.  Levees are constructed on top of the unmapped historical alluvium which continues to be eroded by floods” (p. 287).  He believes this problem is a result of a disconnect between modern process-based geomorphological and river 4engineering research and historical geomorphic understanding.  James (1999) argues that even though historical information is bound to be “fuzzy” it may “provide constraints and insights into process and rates that may prevent the commission of major blunders,”  because “placing modern processes into a long-term context requires knowledge of past process rates and changes which should be validated by historical data.”  The tension in geomorphology between process based and historical studies  is identified by Baker (1988), who calls these themes geomorphotechnics – The “use of scientific and engineering methodology to acquire, interpret, and apply knowledge of the earth’s landscape and the processes operating upon it” –  and geomorphogeny – “the study of the origin, development and changes in the landscapes and landforms of Earth and Earth-like planets” (p. 1160), respectively.  Many geomorphotechnic studies occur at timescales that are much shorter than the long timescales that are typically necessary for significant natural geomorphological change or even the moderate timescales associated with responses to major anthropogenic geomorphic change.  Because geophysical event sequences  exhibit trends, persistence, and intermittency (Church 1980), it is probable that short- term observations of geomorphic patterns will generate to “a misleading impression of the long term variability of a process” (Church 1980). The presence of extrinsic forcing mechanisms further confounds this picture, as current configurations may not be related to the processes that formed them.  It is, therefore, desirable to understand the historical context of systems in which process studies are occurring as this both increases our understanding of perturbations to the system and the length of the geophysical event sequence of interest, even if only qualitatively. In particular, in considering the major effect that humans have had on geomorphic systems, it is critical to understand the history of human interaction with the system.  Petts (1989) also argues that historical analysis of fluvial hydrosystems is of value for understanding system dynamics (as considered above) but, additionally, is critical for river management today as it may be used : “1) To reconstruct the hydrosystems of the pre-industrial era, 2) To determine the rate and magnitude of change of hydrosystems and their components to single and composite impacts; and, 3) To enable the improved prediction of the effects of new impacts and development of ecologically sound management tools” (p. 11).  Human impacts on geomorphic systems have been characterized in various ways. Generally these classifications are based on whether the impact was predominantly on process or form and, somewhat relatedly, whether the result was  intentional or unintentional.  Jones (2001) categorizes impacts as purposeful, incidental, and accidental “to reflect the extent to which the shape of the resulting landform had been predetermined or was merely a by-product of activity” (p. 74). Within the incidental and accidental categories, he distinguishes further between human induced landforms that are created by natural processes “but in location and at times that are wholly dependent on human activity” – such as an alluvial fan into an impounded reservoir– and human modified landforms that are “created when the distribution and rate of operation of geomorphological processes are changed as a consequence of human activity” (p. 76). Goudie (2000) also distinguishes 5between anthropogenic landforms and anthropogenic processes.  He further divides these anthropogenic processes into directly and indirectly modified processes –following Haigh’s (1978) classification. Placer Mining Impacts on Geomorphic Systems  One of the driving forces in the development of historical geomorphology has been the increasingly recognized magnitude of  human impacts on geomorphic systems over the past several thousand years. Higgitt and Lee (2001) state that, “the persistence of human impact as a factor in explaining geomorphological process activity during the last millennium indicates the intricate links between the operation of natural processes and the history of landscape development.” In reflection, they conclude that, “dialogue with historical geographers, historians, and industrial archaeologists offers the prospect of improved understanding of the combined effect of anthropogenic and geomorphological changes and of the reactions, if any, of human settlement to environmental conditions” (p. 280).  Brunsden (2001) states, in the context of an essay on the geomorphic history of Britain, that human activity, “must be treated as the main perturbation to the stability of the geomorphological system during the last 1000 years” (p. 40).  The degree to which this statement can be globalized is debatable.   Two volumes that extensively treat the topic of human impacts on fluvial geomorphology in Europe are Petts et al. (1989) and Higgit and Lee (2001).  Many have chronicled some of the vast effects of human activity in a more global scope. Some of these include Marsh (1874),  Nir (1983) –who also provides an excellent review of the topic as treated over the first half of the 20th century–  Turner et al. (1990), Goudie (1995, 2000), Simmons (1996), and Roberts (1998).  Whitney (1994) chronicles the ecological transformation resulting from the human presence in North America from 1500. Some of the most intense modifications or erosion rates have occurred in restricted environments and localities, including forest roads, cultivated fields, construction sites, and, of course, mining spoil banks (Douglass 1990).  Mining activities mobilize large volumes of sediment that may enter into fluvial systems. As an example of the magnitude of mining activities, Jones (2001) summarizes some of the impacts in Great Britain.  He reports that Sherlock (1931) estimated that by 1913 humans had moved around 15 billion m3 of material from mines in Great Britain.  He further estimates annual earth-moving activity associated with extraction of industrial and construction materials at roughly 365 million m3 a-1 in the mid 1980s.  Placer mining uses flowing water to mobilize large volumes of sediment near streams and rivers that frequently enter into the water bodies.  Nir (1983, p. 67) states that the practice of placer washing  “creates an entirely artificial flow regime of discharges and velocities probably never attained by a natural stream.  The volume of removed alluvia and other clastic material is, no doubt, considerable .” Kennedy (2009) describes the local affect of various placer mining techniques. Traditional sluicing, ground sluicing, and hydraulic mining have had the most substantial lasting 6Figure 1.1:  Typical sluice box (A), ground sluice (B), and hydraulicked (C) placer sites. Site A is Cameron’s Flat, B is the south site on the right bank just below Siwhe Creek, and C is an unnamed hydraulic mine above the Quesnel River near Sardine Flat. 7effects on the landscape.  All three of these technologies involve the use of flowing water to mobilize and sort sediment. Smaller fractions (generally gravel and smaller) of the waste are washed away from the site, often directly into the river, sometimes into tailings piles. Lag deposits (usually cobbles and boulders) are left behind. At sites associated with the use of sluice boxes (figure 1.1A) the ground surface is often lowered by 1-3 m and lag gravels are often left in neatly stacked rows. Ground sluicing (figure 1.1B) leaves eroded depressions with steep scarps and level floors, and large heaps of cobbles. Hydraulic mining (figure 1.1C) leaves “extremely high steep scarps and gently sloping wash pits littered with large boulders, cobbles.” Sedimentation in downstream water bodies is a common result of both placer and lode mining activity if tailings are not carefully managed.  Several examples of impacts on fluvial systems from mining follow.  Knighton (1988) describes 40 million m3 of mining debris resulting from the mining of alluvial tin that was supplied to the Ringarooma river resulting in aggradation, development of a braided pattern, and a 300 percent increase in channel width.  These changes were followed by degradation that began upstream and progressed downstream, reaching rates of 0.5m a-1, and channelization into a single thread.  Knighton predicts that it will take ~150 years for the stream to “cleanse the channel of mining debris.”  Marron (1992) reports that 69 million cubic meters of tailings from lode gold mining were dumped into the Belle Fourche river system after a period of unquantified placer mining in the region.  She determined that about 29 percent of this sediment is deposited along the river’s floodplain and about 60 percent of the sediment is contained in over bank deposits.   Deposition along the river resulted in rapid channel migration.  Placer mining in California produced a very large amount of sediment that entered into the Sacramento and tributary rivers.  James (1997, citing Heuer, 1891)) reports that records of water use were used by Banyaurd (1891) to estimate over 1 billion m3 of sediment were produced. Gilbert (1917) assumed that this figure was determined by collecting records of water use and assigning an approximate duty of gravel moved by mining per unit volume of water (referred to as a miners inch; in this case probably actually a miners inch day) but the range of duties is from 1.5 to 36 yd3 per miners inch.  Gilbert (1917) undertook a surveying effort to independently estimate the total volume of gravel produced.  He manually surveyed hydraulic pits and believed that the surveyed volumes were “true within 10%” (p. 39). Gilbert found that his estimate of the total value exceeded earlier estimates that were based solley on the historical water usage by 51% and presented a summary figure of 1.3 billion cubic meters of sediment produced by mining on tributaries of the Sacramento River.  This sediment rapidly moved downstream causing several meters of aggradation and resultant destruction of farmland along the Sacramento River.  This destruction led to an injunction against placer mining in California 1884.  Degradation of channels followed the aggradation (Gilbert 1917) though large amounts of sediment remain stored in overbank deposits causing a persistent impact on the river system (James 1997). A second phase of licensed hydraulic mining occurred from 1893 to 1953 and resulted production of at least 24 million m3 of sediment.  This estimate is a minimum 8because it is based on the storage of sediment in reservoirs with small trap efficiencies (James 1997).  Rohe (1986) summarized more generally placer mining’s impact across the western United States. Although dominated by the activity on tributaries of the Sacramento River in California, there were many other placer regions in the western United States including southwestern Montana, central Idaho, northwestern California and southwestern Oregon, northeastern Oregon, and north central Colorado. A more comprehensive account of mining in the United States is provided by Koshmann and Bergendahl (1968).  Mining had an extensive impact on vegetation as forests around the mines were cleared to supply building material.  Another author (Dasmann 1999) notes that placer mining in California accelerated colonial settlement and its environmental consequences including grazing, draining of wetlands, logging, and hunting.  Rohe (1986) also notes impacts on upslope water bodies due to diversion and the construction of reservoirs. The greatest direct and district impacts of mining were geomorphic.  Rohe contends that “traditional methods” of mining including panning, rocker box work, and sluicing had a very small scale, but widespread disturbance.  He acknowledges, however, that no statistics exist regarding the extent of the impact of such activity.  Hydraulicking excavated huge amounts of material creating vast amphitheaters.  Sediment discharged from these hydraulic mines had massive impacts on downstream water bodies causing aggradation, channel instability, and inundation of low lying areas.  Estimates of the geomorphic impact of placer mining in British Columbia have not been previously made.  Galois (1970) addressed the general environmental impact of placer mining in the Cariboo.  Bulletin no. 21 of the British Columbia Department of Mines (1946) lists principal placer mining areas in the province, which are listed in Table 1.1. Table 1.1: Key placer gold regions in british columbia.  Asterisks (*) indicate regions of known hydraulic activity.  Data are from British Columbia Department of Mines (1946). Value Okanagin 9 The gold rush on the South Island of New Zealand included significant groundsluicing, hydraulic mining, and extensive dredging (Baker 2001).  A total of  825,353 kg of gold were produced on the South Island between 1857 and 1960 (Salmon, 1973).  Nir (1983) used the reported production yields of six companies, who, combined, produced 1,820 kg of gold from 358,750 tonnes of gold- bearing deposits to estimate that, on average, one ton of deposit yielded 5.1 grams of gold.  This leads to an estimate of approximately 160 million tonnes of material (~80 million cubic meters, bulk volume) having been moved by mining.  Specific estimates for one river system on the South Island, however, are greater than this summary figure.  Placer gold mining in the Clutha Catchment on the South Island of New Zealand began in 1861 and continued into the first half of the 20th century.  The Otago Regional Council (2008) reports that the Rivers Commission (1920) estimated 300 million cubic yards (229 million cubic meters) of material were moved in the Clutha Catchment by placer mining.  They estimated that, as of 1920, 40 million cubic yards had washed through the system, 60 million cubic yards were actively interacting with the river, and 200 million cubic yards had not entered into the river. According to the Rivers Commission, aggradation on the order of 3 m, occurred as a result of this sediment input.  Otago Regional Council (2008), on the basis of repeat cross section measurements of the river, reported degradation starting in the 1920s, at the cessation of placer mining in the region.  This degradation has had a magnitude up to 5 m and has propagated downstream through the 20th century.  The Otago Regional Council Report(2008, p. 25) states that “the general trend of degradation working downstream from Roxburough from this time is most likely related to this reduction in bed material supply to the river,” though, since the 1954 closure of a dam at Roxburough, degradation has probably been driven by a very large reduction in sediment supplied to the Lower Clutha.  The examples of the impact of placer mining outlined above support the notion that human modification of the landscape associated with mining (especially placer mining) has been a key geomorphic perturbation in many regions of the world in recent time (eg. Brunsden (2001). Furthermore, the case studies provide some good examples of sluglike sediment movement through river systems following placer mining Methodology for Using Historical Sources in Environmental Science  Though it is clear that there is much potential value in using historical sources to document both natural and anthropogenic change in the physical environment over timescales of 100 to 1000 years, as has been discussed above, relatively little theoretical work has been done to clarify appropriate methodologies for such work.  In the most comprehensive work on the subject, Hooke and Cain (1982), propose the following sequence of tasks for a researcher who is using historical sources to study the physical environment: 1) Define the purpose and scope of the study. 2) Identify possible sources and investigate their location, availability, characteristics and content. 3) Search for relevant material within potentially suitable documents. 4) Check general accuracy and 10 reliability of documents. 5)Make a pilot study to test the proposed framework for data abstraction. 6)  Abstract information from data sources. 7) Corroborate all information obtained from historical documents. 8) Process data using procedures such as standardization and quantification. 9) Analyze and interpret the data.  This operational sequence provides a useful framework for a review of the work that has been done to inform historically based inquiry about the environment.  This work includes both specific guides for environmental scientists in applying historical methodologies, and the base literature in historical geography that is concerned more generally with human interaction with the landscape.   The succeeding paragraphs will consider: 1) sources of historical information, 2) assessment of reliability and accuracy, 3) abstraction of useful data, 4) corroboration and verification of sources and data, and finally, 5) processing and analyzing the data. Sources of Historical Information  Many documents provide extensive lists of historical sources that may be useful in environmental inquiry.  The range of potentially useful sources is large and “generalization about sources can be difficult in view of the ingenuity of environmental scientists seeking datable evidence” (Richards 1984).  Hooke and Cain (1982) list many sources for historical information in Britain.  Trimble and Cooke (1991) and Trimble (2008) describe sources in the United States. Examples of such sources include estate, tithe, county, cadestral surveys, land use survey maps and marine charts; historical aerial and ground based photography, printed topographic descriptions, newspaper accounts of extraordinary events, government reports, travel and exploration memoirs, meteorological and hydrological records, engineering plans for bridge and canal construction and modification, taxation records, and company log books.  Furthermore, Trimble (2008) emphasizes a whole class of archaeological indicators, including erosion or deposition recorded by the current stream bed elevation relative to dams, mills, fords, and fish traps; movement of geomorphic features relative to roads, canals, causeways, and buildings; and burial of litter and other human artifacts.  Of course the broad range of dating techniques –such as radiocarbon, cosmogenic radionuclide, tree- ring, and lichonometry– that do not rely on human interaction with the environment and are more typically employed by environmental scientists are also useful and may be used in conjunction with historical records in a variety of ways (for example see discussion of corroboration below).  Once records have been identified that may contain useful information, they must be carefully scrutinized. This may require much sifting through unwanted material (Hooke and Cain 1982) in order to find actually relevant data.  Historical sources can be classified in a variety of ways. A first order classification would divide archaeological from text and map based records.  These may be categorized in different ways. Familiar typologies classify them according to source (as primary or secondary), acknowledged bias (as objective reporting or subjective evaluation), and form (as written, printed, or oral).  Historical sources may also be classified by the type of source, which emphasizes the influence of source on a document’s intention, tone, and coverage (Little 1992). 11 Reliability and Accuracy  Hooke and Cain (1982) identify several sources of error in historical documentation.  Error may be introduced during data collection as a result of the frequency and timing of observations, the perception and attitudes of observers, the style and conventions of writing (with special attention to the probability of exaggeration), the semantics of descriptions, and the methods and instruments used in collecting quantitative data.  Trimble (2008) underscores that in assessing the perception and attitudes of observers both “the scientific credentials of the observer” and “the stage of scientific development at the time of observation” must be considered. Furthermore, the temporal and spatial  (Gurnell et al. 2003), and societal representativeness (Butlin 1993) of observations must be considered. Butlin (1993) states that there is a tendency for records of rural and urban and industrial ownership and production to reflect only the experience of the privileged owner-classes and thus to reflect only part of the experience of a whole society at particular times.  In reconstructions of human impacts on the environment this bias could easily translate into blindness towards distributed high frequency-low magnitude impacts of people in non privileged classes.  Examples of such impacts include agricultural activity and deforestation associated with the gathering of fuel wood.   It is also important to recognize segment breaks in the record associated with individual lifespans (Fisher 2002) and changes in the areal units for which data were collected, the systems of dates, time keeping conventions, and units of measurement (Butlin 1993) and the potential for error associated with failure to recognize such changes.  Further sources of error are associated with the reproduction and alteration of original materials during storage.  During compilation and copying, numbers are especially prone to being incorrectly copied, and photoreproduction of maps may produce distortion.  During storage, paper shrinkage of maps may cause systemic and local distortion (Hooke and Cain 1982).  Several internal checks for reliability should be made (Table 1.2). Once the general reliability of a source has been established,  the accuracy of individual statements should still be scrutinized by Table 1.2: Internal checks of reliability (after Hooke and Cain Table 3.2, p. 73) 12 external corroboration. Abstraction and Interpretation of Historical Data  Typically, before information contained within historical documentation can be of use to to an environmental scientist, it must be abstracted or interpreted in some way.  The specific information to be abstracted will depend on the study, although “information on dates and locations is likely to be required in all studies” (Hooke and Cain 1982 p. 83). Hooke and Cain (1982) caution that it is best not to summarize the contents of a document during transcription, as interpretation is prone to be affected by other source material and later study.  A helpful concept regarding interpretation of historical documents is  the idea of code within historical sources as developed by Harley (1982).  Harley argues that by applying a linguistic theory to historical evidence, its value and use in understanding the way the evidence shaped history could be further developed. Harley (1982) refers to  Jakobson’s (1960) model of communication as being constituted by a message communicated to an addressee by an addresser through a particular mode of contact in a code that is shared by the addresser and addressee and defined by the context of that communication.  In order to understand the meaning of a message this entire context must be understood.  This code may be linguistic or numerical.  In the  second case it may have to do with differences in recording conventions and units as mentioned above.  Quantitative data may be derived from qualitative descriptions through a variety of means such as content analysis, a well established technique in the social sciences (eg. Krippendorff 2004) and the application of proxy records (eg Noble and Tongway 1986).  Fisher 2002 describes an attempt to extract temperature records from qualitative descriptions:  “the historical or documentary sites... are based on such things as diary entries, records of plant behavior, crop yields, freeze-ups, etc. These multiple sources are a challenge to homogenize into a temperature-like record and the spectral veracity of such records is very difficult to assess.”  There are semi-quantitative descriptors in many historical sources such as the phrases “largest flood in living memory” or “unprecedented.” Russell (1982) provides an example of how such records can be combined with quantitative records through a Bayesian model to produce quantitative estimates. Corroboration  Many workers agree that it is critical to corroborate all data obtained from historical sources. In their “steps in compiling historical data” Gurnell et al. (2003) suggest that the last stage should be to “field check changes indicated.”  Hooke and Cain (1982) list as an intermediate step “corroborate all information obtained from historical sources” and are open to the variety of sources that would provide reasonable corroboration.  They suggest six possibilities: 1) Comparison of like historical sources that are contemporaneous, of comparable detail, and different ultimate sources; 2) Comparison of evidence from different types of historical sources such as corroboration of a map by written sources or corroboration of government reports with an individual’s travel journal, 3) 13 Comparison of stable features that would not have changed since the historical document was created with the modern landscape; 4) Field survey (which is closely related to the previous technique); 5) Comparison to modern day process studies to determine if an event described is probable or possible; and 6) Circumstantial evidence and proxy data, which are loose groupings of related data, such as records of farming activities as corroboration of a climatic reconstruction.  It seems reasonable to pursue multiple lines of evidence to support belief in any given past environmental event; however, it also seems important to consider that specific corroboration is not always possible or a useful allocation of resources.  In particular, where a historical description can be confirmed with physical evidence, one should consider what combination of historical and field research will yield maximum knowledge of the past environment.  Whether or not a particular historical description can be corroborated, the data obtained should be treated with some skepticism that is informed by present knowledge of both physical processes and related histories.  Studies which link field evidence with historical documentation have a unique power.  In these cases, the direction of corroboration could either involve field checking inferences drawn from historical sources or vice versa, using historical data to aid in the interpretation of physical evidence by verifying or potentially increasing the detail of  physical reconstructions.  Historical sources can be particularly helpful as a dating tool for establishing the timing of either natural or anthropogenic change (Hooke and Cain 1982).   Butlin (1993) emphasizes the value of linked field and historical study in historical geography, which, by extension, informs the understanding of the use of historical sources in environmental science.  He says, “while much of the research undertaken by historical geographers had had its roots in libraries and archives, part of its excitement is the linkage which can be made between the written sources and the field evidence of the places researched and studied. ... Much valuable study has been and is being undertaken of relict features of the landscape, some obvious from the ground, others initially visible only through remote sensing via air photographs and satellite imagery” (p. 89).   He further cautions that field work alone is prone to overly simplistic interpretation without seeking other available means of evidence. Butlin emphasizes the variety of tools available in the physical sciences that may be of use in landscape reconstruction.  He concludes, from the perspective of a historical geographer, that “the distinct possibility exists for major interdisciplinary research on human interaction with river catchments and regimes, using documentary and environmental evidence, and it would be encouraging to see historical geographers and paleohydrologists taking a lead in such projects” (p108). Data Analysis  Analysis of historical data is not inherently different than analysis of any data about the physical environment.  The particular procedures, of course, will depend on the question being addressed in the study and the available data.  A key is maintaining perspective on the degree of accuracy of the historical sources(s) used in the study.  Even this task however, is not inherently different from any research using data from external sources (Gurnell et al. 2003).  In a review of 14 Hooke and Cain (1982), Richards (1984) points out that there are “serious, specific, methodological issues” that remain to be resolved before researchers can confidently use historical sources to address questions of environmental change over the timescale of 100-1000 years.  The first issue he proposes is the difficulty of using secondary sources in quantitative investigations when the representations in those sources are known to be censored by the observer.  The second is the difficulty of evaluating equilibrium and change in environmental systems with arbitrarily dated historical sources.  Richards cites the example of Ferguson (1977), stating: “The sinuosity of a river reach increases as bends tighten, then reduces sharply when a cut-off occurs. ... If an early map is produced right after a cut- off occurrence and a later just before another, there will appear to be an increase in the sinuosity between the map dates, suggesting (incorrectly) a temporal disequilibrium.”  Another related issue is in understanding appropriate sampling designs for temporal data series and the fact that a researcher has no control over the time series available in historical sources. Richards (1984) fails to acknowledge that these issues are less of a problem where the historical records were produced as a consequence of notable environmental changes; for instance, newspaper records of flooding or government records of mining activity.  He does however, note an “emphasis on the remarkable in documentary sources” that may cause misleading perceptions of the relative roles of gradual change and extreme events. Finally, he emphasizes the difficulty to establish causality based on historical understanding derived only from spatial and temporal coincidence.  General Project Methodology  Because this thesis is concerned with an historic environmental perturbation for which some documentary evidence is available, the methodology for using historical sources in environmental science as presented in the previous section provides a useful starting point for a description of the overall project methodology.  Historical data and descriptions are used to develop an overall picture of the scope and scale of the placer mining impact.  Field-based physical investigations are then used to corroborated and refine the picture of the impact of placer mining as developed from historical sources.  The sequence of tasks proposed by Hooke ad Cain (1982) for historical source-based environmental inquiry provides a useful template for the historical component of this thesis.  The purpose of the project is to understand the scale of the placer mining impact on Fraser River and its tributaries, and to assess the potential for persistent influence of the activity, especially on the river as it flows over its gravel fan below Hope.  Important historical information then, is that which pertains to placer mining activity on or above any sediment tributary of the Fraser, but is most keenly interested in activity along the main stem of the river and in large tributaries where there is probable sediment connectivity on the centennial timescale.  Historical sources are used to refine this definition and limit the scope of detailed study to regions where impacts were substantial.  Chapter two presents a summary of historical information regarding mining along the Fraser.  Possibly valuable historical sources were largely identified by Kennedy (2009, and personal communication 2010) and Galois (2009, personal communication).  Principal useful sources proved 15 to be the Annual Reports to the Minister of Mines, records of the Mining Commissioner’s ledgers, early maps, and early compiled histories of the activity (Bancroft et al. 1887 and Howey 1914). Relevant material within these documents included base information for reconstruction of the 19th century typonymy and mental map of the region (place names and descriptions of locations), legal records of mining activity (water rights, sales of claims, etc), qualitative descriptions of the kind(s) of mining that occurred, and records of the amount of gold produced by mining that provide a temporally and regionally resolvable quantitative indication of the scale of the mining activity.  Data from this material is abstracted in a variety of ways.  Much of it is placed on a medium- scale map to allow geographic contextualization of a qualitative narrative that was also based in this source material (Map 2.1). Other data were abstracted in the development of a table of estimates of the gold produced in regions through time.  The reliability and accuracy of the historical records were checked by evaluation of the internal checks of accuracy listed by Hooke and Cain (1982 table 3.2) and assembling many different and somewhat independent records to allow for historically based corroboration.  Most of the sources used were contemporaneous (the notable exception  being Howey 1914) and posessessed moderate propinquity (most were recorded at regional centers from which a substantial proportion of the mining was quite remote: this leads to a situation where positively recorded information is probably reasonably accuratebut the absence of records does not necessarily correspond to the absence of activity).  The level of generalization in the consulted records is extremely variable.  The records of the Mining Recorders and Commissioners are very detailed, providing specific points of information in space and time, while the annual reports to the Minister of Mines tended to provide general annual summaries of activity over fairly large regions (mining districts or divisions which include tens to hundreds of km or river) although they sometimes include highly detailed reports of the activity of capitalized mines. The Annual Reports are highly generalized regarding non-capitalized “desultory” mining, especially when that mining was undertaken by Chinese or First Nations labor.  The consulted records were typically machine copies of original material and direct quotations of letters written to the Minister of Mines by regional officers in the case of the annual reports. Maps of the region, on the other hand, tended to be published quite remotely (in Victoria, San Francisco, or London), and on the basis of unreferenced sources.  Source material was written for three primary purposes: to record legal rights to resource access, to provide a record of the development of mining activity for the colonial government to aid in maintenance of British/Canadian colonial control over the region, and, in the case of some material that was used primarily for secondary corroboration, to influence people to enter into the mining industry either by traveling to the region (often pamphlets were published by steamship companies) or by providing capital to the industry.  Much placer mining leaves a distinct and persistent physical record in the landscape. Because of this, it is possible to corroborate historical information about the mining with physical evidence in addition to internal material.  Chapters three through six report physical investigation 16 of the landscape.  Chapter three presents efforts to map the physical evidence of mining and to develop an understanding of the relation between mine area and volume.  Air photos, ground-based investigation directed by historical data, and a raft-transect of the river were used to develop a map of the locations, planimetric extent, and style of mines along the Fraser.  The relation between area and volume is developed by presenting detailed surveys of a selected set of mines and regression relations that are used to predict the volume of mines from the area they cover and kind of mining that was done.  In chapter four, this relation used to estimate the total impact of the mining along the river.  Furthermore, this view of the physical evidence left by mining allows  evaluation of the quality of information about mining derived exclusively from historical documentation.   This physically based view of the impact of mining along the river allows some re-evaluation of the historical record revealing some bias and unrecorded information in that record even while confirming much of the overall picture painted by the record.  Chapter four presents study of the grain size distribution of sediment that was affected by mining which, along with overall estimates of the volume and locations of sediment inputs and timing of sediment inputs (chapter 2), provides the basis from which evaluation of the potential for persistence and downstream translation of mining impacts is possible. Furthermore, this quantitative information on the impact of placer mining provides a basis for comparison with and evaluation of the historical methods used to estimate the impact of placer mining that were used at the beginning of the thesis; and to to qualitatively evaluate the historical record and current historical understanding of the environmental impacts of placer mining.  The potential spatial and temporal geomorphic connection between mines distributed along the middle river in the 19th century and the lower Fraser today is addressed in chapter 6, which present some preliminary hydrologic study of the river and a body of literature relating to sediment slugs and their propagation.  This information allows a first-order assessment of the plausibility of the hypothesis that 19th century placer mining is driving observed aggradation of the lowest gravel-bed reaches of the river today.  This research uses the conceptual model of a downstream propagating sediment slug to evaluate potential connection between 19th century mining and the aggrading lower Fraser today.  Where observations are consistent with this conceptual model, the research may provide limited confirmation, but because of the historical nature of the phenomenon, it can not formally be considered a test of the model.  This means that, in the absence of independent observations, the following two conclusions are as definitive as this research will be able to produce: 1) if the qualitative application of the literature relating to sediment slugs suggests that sediment from the mining would have move into into the aggrading reach in the way we hypothesize, both the conceptual model and hypothesis may be correct, and 2) if there is serious discordance between the conceptual model as applied to estimated sediment inputs and observations of aggradation sediment into the aggrading reach either the observations of the geomorphic impact of mining at mine sites, the conceptual model, or the hypothesis, must be incorrect.   It is possible that some observations 17 could provide independent validation of the model. These include identification of lithologies in the aggrading reach that can be directly traced to placer deposits, the absence of aggraded sediment between placer sites and the aggrading reach, and an absence of accumulated sediment immediately below placer sites. 18 CHAPTER 2: HISTORY OF PLACER MINING IN THE FRASER BASIN  Historical documents and scholarship provide information on the timing, extent, intensity, and styles of placer mining along Fraser River.  Such information provides an important foundation for any attempt to reconstruct an environmental history of the impact of mining on the Fraser and a valuable source of insight into the physical remains left by the mining.  This chapter will first consider secondary source material that describes the mining history and will then attempt a more comprehensive environmental history of mining based on primary sources.  The former will emphasize the human history of placer extraction and follow the narrative of discovery and movement through space, while the latter will focus on the industry’s development through space and time will then be combined with estimates of the total gold production to produce estimates of the geomorphic impact of mining through time. Fraser River Gold Mining Landscape  The historical accounts use an extensive and largely extinct mental map of important landmarks and geographic boundaries --including public houses, settlements, “important” mines, creeks and lakes, canyons, bridges, ferries, trails, regions and boundaries between administrative units – and the names by which they were referenced.  Figure 2.1 identifies some of the administrative and general regional names used in the following accounts. Legal records kept by the mining commissioners and correspondence between commissioners (particularly as recorded in the British Columbia Dept. of Mines and Petroluem Resources publication “Annual Report of the Minister of Mine for the Year Ending...” the Department of Mines and Petroleum Resources, hereafter abbreviated ARMM [pub. date]) provide detailed information on the location and timing of mining activity.  It was possible to place many relevant records that were extracted through archival searching on a map of the region.  This section describes a map of historical records of mining activity along the Fraser and Quesnel Rivers (Map 2.1) and the methods that were used to produce the map.  This map also shows locations of physical evidence of mining; the methods for which are documented in Chapter 3.   Production of the map was a collaborative effort of the author of this thesis and Mr. Michael Kennedy, of Lillooet.  Mapped locations of legal records pertaining to placer mining provide good indications of the locations where mining activity occurred.  This information is critical as a base for seeking physical evidence of mining that can be used to reconstructing the environmental impacts of placer mining in the region and for archaeological investigation related to the mining history.  Furthermore, this map of records of mining activity provide a foundational piece of quantitative information on the spatial and temporal extent of placer mining activity along the Fraser.  It shows that the mining industry was important all along the Fraser and that activity occurred nearly continuously from 1858 to the end of the century. In addition, the locations of many names of mines, creeks, and mining areas that are referenced in the following text are shown on the map.  Michael Kennedy and Dr. Robert Galois collected provincial and colonial records relating to 19 placer mining for the period 1857-1910 from the BC Provincial Archives (personal communication). Galois collected records from the High Bar area north to the Cottonwood Canyon and along the Quesnel and Cariboo rivers upstream as far as Quesnel and Cariboo lakes while Kennedy collected records from Big Bar downstream to the edge of mining activity.   The records they discovered took the form of government agency records kept in multi-year leather-bound ledgers, loose sheets/maps, and bundles of Free Miners Certificates (counterfoils).  In-text citations for these sources will list the Government of British Columbia Archives, Victoria BC (BCARS) call number (eg: GR 224 box 12) The most valuable of these were records of claim transfers and records of water privileges and transfers, which often contained anecdotal information on the location of claims and water sources. Table 2.1 lists sources used in the creation of this map.  The government records, which are the key sources used to create the map, represent a view of the mining activity and landscape that is filtered by the Mining Commissioners understanding and does not necessarily reflect its entire scope or the mental map of miners themselves.  This difference is probably especially pronounced where language and/or cultural barriers (as in the case of Chinese and First Nation miners) prevented the adoption of a common toponymy and communication about ongoing mining activity, as evidenced by the gold commissioners’ frequent complaints, documented in the annual reports, of the difficulty to obtain information about Chinese mining activity. U S A  Y A L E D IS T . L IL L O O E T D IST. C A R IB O O D IS T . L IL L O O E T D IS T . Q u es ne ll Fo rks / K ei th le y D iv is io n Quesnellmouth Division C li n to n D iv . L il lo o e t D iv .   F ra s e r R iv e r      Y a le D iv . A sh c r o ft D iv. Big Bar Quesnel Hope Lytton Lillooet Yale Vancouver Williams Lake Boston Bar Aggasiz Mission Barkerville 	                  Quesnell Forks Couteau Country Canoe Country Cariboo Country Early (~1859) Regional Names 	  Fountian 0 25 50km Figure 2.1 General locations of administrative boundaries, region names, and some of the communi- ties referenced in the following text.  Boundaries between these areas are somewhat fuzzy and vari- able in different time periods. Mining districts boundaries are shown as long-dashed lines.  Districts are subdivided into divisions, the boundaries of which are shown divided by short-dashed lines. 20 Source Location(s) Type GR-0224 Box 21 1859-1861 Lillooet Mining Records GR-0224 Box 22 1865-1867, 1887-1896 Lillooet Mining Records GR-0252 Box 12 1858-1872 Yale Mining Records GR-0252 Box 12a 1858-1866 Yale Mining Records GR-0252 Box 13 1873-1888 Manual of Record GR-0252 Box 14 1880-1890 Manual of Record GR-1054 Box 1 Folder 1 1859-1874 Mining Records GR 0216 vol. 30-33 1860-1862 Alexandria Mining Records GR 0216 Vol. 76 1867-1899 Mining Records 1874-1910 All Mining Reports 1862 Middle Fraser Map Bowman 1887 Map 1862 All Map Nation 1910 Map Ward and Harris 2001 (a & b) All Map Collection Various, Recent All Map Series 2011 All Bancroft et al 1887 All Published Narrative 1926 All Published Narrative 1914 All Published Narrative 1858 Mid-Lower Fraser Published Narrative 1858-1859 Mid-Lower Fraser Newspaper 1858-1859 Mid-Lower Fraser Newspaper 1858-1859 Mid-Lower Fraser Newspaper 1858 Mid-Lower Fraser Newspaper Record Time Period or Year Published Lytton and Yale Lytton and Yale Lytton Quesnellemouth Annual Reports to the Minister of Mines Conroy Cariboo (including Quesnel River) Epner Lillooet and Clinton dvs. Canadian Topographic Maps 1:50,000 GeoBC Geographic Names Database Howay Howay Waddington Victoria Gazatte San Fransisco Evening Bulleting Alta California (San Fransisco) Northern Light (Whatcom) Table 2.1: Sources used in the creation of Map 2.1. 21  Because the toponymy used in these records is largely extinct from common use and Geographic BC names database (GeoBC 2011), early maps that used the toponymy in the Gold Commissioners records and that could be correlated with the current landscape provided a very valuable base for the mapping effort.   Particularly useful early maps included (Conroy 1862, Epner 1862, Bowman 1887, Nation, 1910, and those in the collection of Ward and Harris (2001 a & b).  Once a skeleton map of the nineteenth century mining toponymy was in place, it was possible to iteratively process and cross reference records of mine locations and other geographic features.  Each pass produced an increasingly complete picture of the commissioner’s mental map of the mining activity and regions that contained it.  Anecdotal descriptions of mine locations were often similar in form to the following: “on the left bank of Fraser river about 7 miles below Quesnel” (GR 216 vol. 76).  This often provided sufficient detail to narrow the area of the mine to within approximately two kilometers of one side of the river.  Symbols representing mine locations were then placed at the most probable location for the mine, considering locations of likely placer gold accumulation.  Possible locations were constrained by the presence of fluvial gravel and often included gravel that was near to bedrock or “false bedrock,” downstream of obstructions, or terraces in areas where the river is mostly constrained by bedrock. Initial Discovery and the Gold Rushes  Historical accounts of placer mining along the Fraser and its tributaries typically emphasize the “Gold Rush” element of the history to the exclusion of detailed discussion of ongoing industrial activity through the beginning of the 20th century.  In Ormsby’s (1971) history of British Columbia, some account of the Gold Rush history is provided, but emphasis is placed on the way the influx of miners caused changes in governance and forced the creation of transportation infrastructure. Because the emphasis is on how gold mining caused societal change, Ormsby emphasizes the early history of the mining, when new areas were opened to colonial settlement, and describes the initial gold rush of 1858, expansion of the gold mining effort up-river in 1859 and 1860, and discovery and development of the Cariboo Gold Fields from 1861 to 1863.  She does not describe activity continuing past 1859 on the lower part of the river near Yale; leaving the story of mining on the lower river with, “in 1859 the old bars were practically abandoned to the patient Chinese as the other miners moved north to the district above Lytton.”  Bulletin No. 21 of the British Columbia Department of Mines (1946) provides a brief, but broader, account of the mining history.  This report summarizes the extent of mining activity along the river but includes little detail regarding the timing and intensity of mining operations.  According to the report, the lowest reported gold bearing bar1 on the Fraser that was profitably worked was Maria Bar, 25 miles (40 km) below Hope.  Ascending upstream, mention is made of Hill’s Bar, which “probably yielded more gold than any other single locality on the Fraser ...at least $2,000,0002 worth” (p. 22).  Runs of “heavy” gold are reported from a few miles below Boston Bar to Sisco (probably 22 today’s Siska) Flat, below Lytton, and from Fosters Bar below Lillooet to “some little distance above Fountain.” Still further up-stream, bar and bench diggings extend “past Alexandria and the mouth of the Quesnel River to Cottonwood Canyon.” (p. 23).  Regarding timing, the Bulletin states that many of the placer deposits were “found and worked in the first years after the initial discovery” and that by 1875 “most of the miners were Chinese and Indians” (p. 23), presumably indicating that  relatively low-paying deposits were being worked.  The bulletin provides some information on  the styles of mining: “the early bar deposits were shallow and were worked with pans and rockers. Subsequently, some of the benches where hydraulicked and and several dredges were built.”  The most comprehensive secondary source about placer mining along the Fraser is Howay (1914).  A summary of his writing, with occasional interjections from other sources, about the discovery and development of placer mining will hopefully serve to contextualize the primary sources that will be discussed later:  Parallel arguments suggest that Howay may have used the hisotry of Bancroft’s et al. (1887) as a key but unacknowledged source.  Governor Douglas considered the first discovery of gold on the Fraser system to have been made on the Thompson, a quarter of a mile below Nicoamen in 1856 (qtd in Howay 1914). By 1857, prospectors had moved from the Columbia to the discovery on the Thompson (p. 12).  A group of miners ascending to the Thompson by way of the Fraser discovered rich pay on Hill’s Bar in March, 1858.  By April, news of the discovery reached San Fransisco and the the miners from the ‘49 rush who remained there.  The California mining industry had been transformed during the 1850s from one dominated by individuals and small companies, with estimated wages of $16 to the man per day in 1849, to organization in companies employing wage laborers with wages of $3  to the man per day in the late fifties and early sixties (Jung, 1999).  Miners flocked to British Columbia from theese relatively depressed conditions in California.  On April 25th, 450 Californians arrived in Victoria aboard an American steamer, The Commodore (Ormsby 1971 p. 139).  Miners continued to arrive through the summer. An estimated 25,000 people landed in Victoria during 1858 (Ormsby 1971 p. 140). Presumably, not all of these people were directly engaged in mining; many provided support services.  In late April, 1858, one miner, Franklin Matthias reported in a letter to a friend that he is a member of the highest party on the Fraser “thirty miles above the Junction of the Fraser and Thompson Rivers” and that they intend to move above the Falls (Bridge River Rapids) once enough men were gathered so that they would not be intimidated by the Indians.  Matthias reported, “Almost every bar on the river from Fort Yale up will pay from $3 to $7 a day to the man” and that on some of the bars above Lytton people were making up to $16 a day each (qtd in Howay 1914 p. 19). By June miners were working as far north as “the Fountain” above Lillooet.  High water of the June freshet delayed bar mining until late August, causing thousands of would-be miners to give up hope and leave the country “under the conviction that the water would never fall sufficiently” (Bancroft et al. 1887 p. 440). In late August “two hundred men were at work on Fort Yale Bar, and four hundred on Hill’s Bar.  Like peaks above the deluge, the bars began to appear with the gradual recession and were greedily seized upon by the waiting hundreds.  Miners were now scattered all along the river 23 from Fort Langley to Pavilion” (Howay 1914 p. 38). By October ten thousand miners were officially estimated to be working on the Fraser (Douglas qtd in Bancroft et al. 1887 p. 442).  During this very early phase of mining, most mention is made of working river bars with rockers, although at least one report mentions that “we have sluices, also” (S. Allen qtd in Howay 1914 p. 21).  Howay (1914) reports that rockers were soon replaced by sluices, which required substantial capital investment and/ or division of labor for milling lumber that was used to make sluices, digging and managing ditches to provide water, and the actual working of mines. As early as summer, 1858, bar diggings were being supplanted by bench diggings (p. 41).  By November 1858 there were at least thirteen ditches operating between Cornish Bar and Hope (Bancroft et al. 1887 p. 443).  Ditches convey water from creeks to “dry diggings” on terraces above the river.  The presence of ditches indicates that these diggings were worked by sluices or ground sluices. One prospector, Aaron Post, explored upriver as high as the Chilcotin confluence in 1858, and reported finding gold on every bar (Howay 1914 p. 69). During the season of 1858, the core of mining activity remained between Hope and Yale. At the close of the 1858 mining season, many of the Californian miners returned to San Francisco while others overwintered in Victoria (Ormsby 1971.).  Late in 1858 a trail and water route was opened to Lillooet from Port Douglas at the head of Harrison Lake which assisted movement of individuals and supplies to sites higher on the river. In the spring of 1859, individual miners without access to capital pressed  upstream while, lower on the river, most of the bars in the river were deserted for dry (sometimes referred to as bench) diggings where substantial co-operation and/or capital was required to dig ditches to convey water to  sluices. At the same time, Ormsby emphasizes regional movement, remarking only that the white miners moved upriver abandoning bars near Yale to the “patient Chinese” (1971, p. 181).  The Fraser was prospected as high as Fort George and the center of mining moved upstream to somewhere near Lillooet. Howay states that “Between Alexandria, Fort  George, and Quesnel Lake in 1859 about one thousand men were mining on the various bars.”  Even high on the river at British Bar and Rich Bar just south of the confluence of the Fraser and Quesnel Rivers, sluices were in operation in 1859 (p. 72).  The period from 1858-1859 probably was the peak for gold production on the Fraser.  In 1860 “Doc” Keithley and companions discovered rich ground on Keithley Creek, a tributary of Cariboo Lake.  This and other reports from the Quesnel country resulted in a dramatic reduction in the number of people working in the Yale district, with only 203 people working in the vicinity of Hope and earning from $3 to $12 per day (Howay 1914 p. 72).  Following this, many auriferous creeks of the Cariboo were discovered in 1860-61.  Development of placer resources and road infrastructure necessary to support the mining population continued.  The year 1863 “was the banner year of Cariboo... gold was being produced on a scale which exceeded California in its palmiest days” (p. 110) and  production remained high in 1864-65.  By 1865, however, most of the production was coming from a few rich claims, leaving many former prospectors working as wage earners.  Although mining was focused in the Cariboo in the sixties, it continued along the main stem 24 of Fraser River.  A letter quoted in Oliver’s Handbook (1862) describes mining on the benches of the river and indicates that hydraulic mining was taking place in the region: “Beside gold found in the beds and on the shores of these streams, the Fraser itself and many of its tributaries are skirted or bordered by terraces, all of which yield gold also.  These terraces, or ‘benches,’ as the miners call them, run, at intervals, along both sides of the rivers for miles in length; and they recede , where the mountains retire, for distances back into the valleys, varying from a few acres to a few miles in breadth. ... These terraces are composed of the ordinary alluvial deposits – loam, gravel stone, sand and boulders... they contain vast deposits of gold; and to be worked to advantage, the ‘bench digging’ must command a stream of water supplied from a source higher than their own surfaces, so as to give a fall to enable the miner to apply the water to the face of the ‘bench’ by a hose. The force of the stream is due to the height of the fall.  A good strong stream playing upon the face of the hill will disintegrate a great quantity of ‘pay dirt’ in a short time.  The floating rubbish or ‘dirt,’ is caught in a long sluice at the base, provided with ‘riffles’ on the bottom, and spread with quicksilver to catch the gold.  This mode of mining is called by the miners, ‘hydraulic mining’” (p. 57).  Howay reports little of the mining history between 1865 and 1875. He mentions substantial hydraulic operations on Williams Creek of the Cariboo in 1875 and a good recovery of gold in 1876 from the Quesnel River due to very low water conditions, indicating that a substantial number of small scale miners were working there at that time.  Though his account is the most comprehensive that is available, Howay does not consider the history of mining activity in the Cariboo after 1876 or along the Fraser after 1859.  Kennedy (2009) recounts some of the later history, focusing on the “middle Fraser” between Lytton and Big Bar.  He notes that little documentary evidence is available about mining along the Fraser between the end of the Gold Commissioner’s entries in 1860 and the first annual reports of the Minister of Mines in 1874.  Beginning in 1874, Kennedy notes that “in any given year some six hundred people were mining between Lytton and the mouth of the Chilcotin”; that these miners were mostly of about equal parts native and Chinese miners; and that some of the work being done was on a very large scale.  In Kennedy’s account, widespread hydraulic mining entered the region in the mid- 1880s with important activity at Lillooet, Texas Creek, Fountain, and other localities.  Tree core data from Brownings Flat, an extremely large sluice/low head hydraulic operation yield establishment dates in the mid to late 1880s (Kennedy, personal communication 2010), indicating that mining at that location had ceased by, at the latest, the mid 1880s. Gold discovery on Cayoosh Creek in 1894 temporarily increased the amount of mining near Lillooet.  Gold dredging began near Big Bar, and several operations worked the river over the next two decades.  Kennedy concludes  that “In the first decade of the twentieth century, to all  intents and purposes the last decade of placer mining along the middle Fraser, almost the full range of placer mining technology was in play: dredging near Lytton, Lillooet, and Big Bar; hydraulic mining at Big Bar, Fountain, and Lillooet, short lived attempts at wing damming at the mouth of the Bridge River; and small operations still dependent on pan, rocker, and sluice.” By the end of the decade, nearly all mining had ceased on the middle Fraser. 25 Development of the Mining Industry Along Fraser River Yale Vicinity: Maria Bar to the Hell’s Gate Canyon  The history of the initial Fraser Gold rush, centered at Yale, has been fairly comprehensively described by Ormsby (1971) and Howay (1914).  Of the approximately twenty thousand individuals who arrived in Victoria during the spring and summer of 1858, reports indicate that a substantial number did mine along the Fraser.  Governor Douglass reported that ten thousand persons were engaged in mining along the Fraser, with five thousand between Murderer’s Bar just south of Hope and Fort Yale (qtd in Howay 1914 p. 41).  These figures must be treated with a degree of caution because it is possible that Douglas could have had political reasons to inflate the figures3.  The army of miners who worked the river as it fell in the fall of 1858 employed both rocker boxes and sluices. Rocker boxes had limited potential to modify the landscape; one man operating a rocker box can process up to 3 cubic yards of gravel per day (British Columbia Department of Mines 1946)4.   Even with this volume of gravel processed, the material was not necessarily displaced a great distance. Sluices, on the other hand, have the potential to displace large volumes of gravel. In a sluice, gravel is shoveled into a long box through which water is flowing and the bottom of which is fitted with riffles to retain gold.  Space was required at the tail of the sluice box for disposal of tailings.  One convenient way to dispose of tailings was to  dump them directly into a larger, fast flowing body of water. The size and capacity of a sluice is dependent on the scale of the operation, character of gold, and amount of water available, (British Columbia Department of Mines 1946 p. 29), but Kennedy (2009) reports that with a normal sluice of that era “two people can process twenty to thirty yards of gravel in a ten hour day.”   Many of the earliest reports of mining along the river include reference to sluicing.  Water records from 1858 also indicate that many creeks were flumed to mine sites for sluicing purposes (GR 252 Vol 12 File 1).  Much of the initial work was done with rocker boxes below the high water line on bars of the river, which would have caused a temporary disturbance of the surface but would not have resulted in long-term changes to the amount of sediment available to the river.  However, mining quickly moved to sluicing operations above the high water line that would have resulted in substantial additions of sediment to the river.  It is not possible to discern what proportion of the miners in the summer and fall of 1858 were working on the bars of the river versus its banks and benches, but there was certainly the potential for major landscape modification even in the first year of mining.  With the onset of the winter of 1858-59, “all but about 3000” of the miners who had come into the area in 1858 returned to San Fransisco (Dawson 1889 p. 19R).  Miners also moved farther upstream as gold deposits were found, communication routes improved, and the cost of goods decreased.   With excitement focused higher on the river, and eventually, in the Cariboo, information regarding mining in the vicinity of Yale is harder to locate.  In the spring of 1860, the Hope district was occupied by over 200 miners (Bancroft et al. 1887 p. 444). The movement away from the area concerned Mr. Sanders, the Gold Commissioner at Yale, who in April reported “that he feared the rush into [the Cariboo] would depopulate Yale district (qtd in Howay 1913 p. 72).   Mining continued 26 in the region, however, as records of water rights, sales of claims, and mining activity persist through the 1860s (GR 252 Vol 12).  In 1861, according to Bancroft, two thousand Chinese miners worked in the vicinity of Yale (Bancroft et al. 1887 p. 444).  The first years of  the Annual Reports to the Minister of Mines (1874-1876) do not mention activity in the Vicinity of Yale.  In 1877 there is a report of a group a Chinese miners discovering remunerative diggings on Trafalgar Flat, above Hope, and mention of continuing activity at a claim at the head of Hill’s Bar.  In 1878 there is reference to continued mining on Hill’s Bar flat and other, general activity in the region.   The next year’s reports contain similar references to relatively small scale operations and continued “desultory”5 mining.  The first indication of a substantial increase in mining activity in the region occurs in 1881 with discovery of gold on Saw Mill/Hills Bar Flat to which a large amount of water was brought. By the next year it was clear that the mine did not produce “good results.” Through the rest of the 1880s the Annual Reports indicate that most of the mining in the region was done by small groups of Chinese laborers.  The 1885 report states that there is “considerable mining along the banks of Fraser River” (ARMM 1885 p. 497).  In 1890, another attempt to develop Saw Mill Flat on a large-scale was made which, as the previous attempt, failed: “expectations not having been realized, work was discontinued for the season, after a large outlay for machinery and labor” (ARMM 1891 p. 575).  In 1892 there is mention of a decline in desultory mining along the river.  However, several capital intensive and very geomorphically important developments began in 1893.  The Annual Reports state: “considerable attention is being taken in (sic) the extensive auriferous gravel benches which flank the Fraser River for gravel mining operations and the outlook for the coming summer is decidedly of an encouraging character, judging from the large number of applications received” (ARMM 1893 p. 1070).  Specific mention is made of hydraulic mining at Emory/Prince Albert Flat where 3,500 miner’s inches6 of water were used to make excavations from “ten to thirty feet deep” (ARMM 1893 p. 1071).  In 1894 specific mention is made of hydraulic work on Hills Bar (Sawmill) flat and at Texas Lake.  In the years 1883 and 1884, the Yale Gold Dredging Company worked the river at Hill’s Bar.  The period of intense hydraulic activity near Yale appears to have been in major decline by 1898, when the Annual Report states that “Hydraulic mining seems to have been pretty much at a standstill during the past season,” (p. 1108) though the following year there is mention of development of a new hydraulic lease by the Dewdney Syndicate.   Also, in 1899 the Annual Report gives indication of the extent of smaller scale operations: “About 100 Chinese were engaged between Agassiz and Keefer’s. Ten companies were working with sluices, three others with rockers in a desultory manner” (p 744).  In 1902 a total of nine “principally Chinese Placer Workings” are listed and mention is made of 25 miners engaged in “desultory rocking.” After this, the amount of mining in the Yale district continued to decline. By 1905, the only description of mining was “two or three Chinamen at Cat Landing and three or four at MacRae bar represent the once numerous hand placer miners of this devision” (ARMM p. J206). 27 Boston Bar Vicinity  Boston Bar is situated in a run of coarse gold that extends from “a few miles below Boston Bar (or about sixteen miles above Yale) to Sisco Flat, a short way below Lytton, a distance in all of about twenty-five miles” where “rich deposits of ‘heavy’ gold were worked” (Dawson 1889).  Claims on the bars were worked as early as 1858, as indicated by records of water rights and claims from the Assistant Gold Commissioner’s Ledger, Lytton (GR 252 vol. 12), and the report of Bancroft et al. (1887) who note that Boston Bar rose to be a geographical point of note in 1858 (p. 447).  Records of mining activity near this run of heavy gold indicate substantial activity through 1862.  Few records of activity in this area have been located between 1862 and 1879, when the Annual Report to the Minister of Mines states: “the large unexplored flats which were hastily glimpsed at 21 years ago and left for the rich and easily worked bars of the Fraser are now beginning to attract more special attention” and “claims worked last year in the neighborhood of Boston Bar yielded to their owners a fair return.” Specific reference is made to hydraulic activity: “one company of five men have just completed their hydraulic appliances to work a bench claim on an extensive scale” (ARMM 1879 p. 248).  The next reference to activity near Boston Bar comes in the 1891 Annual Report, when “two leases have been applied for adjoining the Boston Bar farm.” A flurry of activity followed in 1892. The North Bend Placer Steward and Co. of Tacoma, Washington applied for two mining leases, and The Ottawa Hydraulic Mining and Milling Company of Ottawa and “parties in Seattle” applied for four mining leases at Boston Bar.  The Annual Reports mention extensive activity on these leases through 1900.  In 1895, the Hagar and Gardener Company made an attempt to develop a hydraulic lease at Eight Mile Creek, north of Boston Bar, which proved to be unprofitable by the next year. Also in 1895 the Wendell Co. developed a hydraulic operation which utilized an old ditch that was re-opened and flumed where required to gain access to a water source.  The presence of such a ditch indicates that early mining operations in the area must have used sluices and illustrates how ditches were commonly adapted by several “generations” of miners.  In 1897 reference is made to the Agnes Hydraulic Mining Co. having an operation at North Bend, opposite Boston Bar.  One of the most successful dredging operations on the Fraser was the Beatty Gold Mining and Dredging Co. that worked Boston Bar in the late 1890s. They operated a shovel-based dredge, in contrast to the suction dredges that consistently clogged with stones (ARMM 1897 p. 616).  No mention is made in the Annual Reports of either the cessation or continuation of hydraulic mining activity in the Vicinity of Boston Bar past 1900. Lytton Vicinity: Fraser River from Eight-Mile Hill to Van Winkle Flat and Thompson River  As already noted, the first discovery of gold in the Fraser watershed was at Nicomen on the Thompson river in 1857.  The Handbook of British Columbia and Emigrants Guide to the Gold Fields, a Colonial propaganda piece published in London in 1862, reports the presence of mining 28 on the bars and sluicing on the banks of the Thompson in “such portions of the Thompson as run through somewhat level ground” (W. Oliver, 1862) .  An early history of mining similar to that near Boston Bar occurred primarily on the bars of the Fraser and Thompson, and is recorded in the period from 1858 through the 1860s in GR 252 vol 12. Clusters of activity in these reports include Kanaka, Lytton, and the Cameron’s Bar area.  Presumably, the intensity of mining at Lytton decreased in the period 1859-1860 as miners moved up into the Cariboo, but mining continued by those who were willing to work moderately paying claims along the river.  Bancroft et al. report that 150 miners worked at Lytton in April of 1858, one hundred working in January 1859, and that in March of 1859 many of the “upwards of three thousand persons” who “had entered the Cascade region... remained round [sic] Lytton” (p. 449-450). Seven hundred miners worked here in September 1860 and through 1864 “several companies were still taking out considerable sums at... Kanaka Creek... and at other points” (p. 450).  Reports from Lytton to the Minister of Mines begin in 1884.  At this time, mining “is confined to Chinese and Indians working on the Fraser and Thompson Rivers” (ARMM 1884 p. 422). Descriptions of this character describe mining near Lytton vicinity until 1892, when the Van Winkle Bar Hydraulic company’s operation over an 800 acre area is described. This company continued work until 1896. In 1897 and 1898 the Reports mention the Ashcroft Gold Mining Co. working a lease one mile below Keefers.  Dredging was also an important activity in the vicinity of Lytton during the late 90’s through 1903, when the Cobledick and other dredges operated, although probably not profitably (Kennedy 2009).  Though hydraulic mining was occurring in the 1890s, smaller-scale operations continued.  In 1896, the Annual Report states that “Keefer’s, Cisco [Siska], Kanaka Bar and Lytton are localities mined principally by Indians with rockers.” In 1898, mention was made of activity by Chinese miners who “in great numbers are working, principally below high water mark, between Hope and Lytton (ARMM 1898 p. 1108).”  In the first years of the 20th century, mining near Lytton was dominated by continued work on the bars and benches of the Fraser and Thompson Rivers by Chinese and Indians with sluices and rockers.  Particularly good pay came in 1903 ,when, “as a consequence of the very low water in the early part of the season... the yield of placer gold from bars, etc. was much better than usual, Indians and Chinese taking out a large quantity of dust with sluices and rockers” (ARMM p. 182).  The next year, however, “placer mining on the Fraser and Thompson rivers has not been carried on as extensively... as formerly” (ARMM 1904 p. 234) and, by 1906, “No placer mining has been engaged in to speak of” (ARMM 1906 p. 177).  The overall proportion in this vicinity of mining on the Thompson River was probably small. Bancroft et al. (1887) note that that, on the whole the Thompson “had a comparatively insignificant mining record after 1858” (p. 458).   They mention that two hundred Chinese worked near the mouth of the river in 1860 and that in 1861, one hundred and fifty miners were working “not far from Tranquille River.” 29 Lillooet Vicinity: From North of Van Winkle Flat to Pavilion  In 1858 “both bank and river mining were in progress between the Forks of the Thompson and the Fountain” (Bancroft et al. 1887 p. 450).  Bancroft lists the important early mining localities in the vicinity of Lillooet: Mormon Bar, Splindulen (probably todays Splintlum) Flat, Cameron Bar, McGoffey Dry-diggings, Foster Bar, Willow Bank, the Great Falls, Robinson’s Bar, French Bar, upper Mormon Bar, and the Fountain.  Two hundred fifty to three hundred miners overwintered in 1858-59 at the confluence of the Bridge River.  By spring 1859, the Douglas Route connecting Port Douglas on Harrison Lake and Lillooet (then known as Cayoosh) had been completed and Lillooet became the entry point for mining on the Fraser from Boston Bar upstream.  The Gold Commissioners at Lillooet’s ledger (GR 0224, box 21), and other government records  (GR 224 box 22, GR 252 Vol 12 File 1,), indicate extensive mining activity in area in the period from 1859- 62.  Kennedy (2009) notes that in 1859, nearly all of the streams along the Fraser within five miles of Lillooet were used for mining, clear evidence of sluicing.   Again, in the Lillooet vicinity, there is little documentary evidence regarding mining between 1862 and the first mention of the area in the Annual Reports of the Minister of Mines in 1876.  Bancroft et al. mention sixty miners working Foster’s Bar in 1865 (p 452).  The first mention in the Annual Reports of activity in the Lillooet area comes in 1876 when it reports the construction work adapting a pre-existing ditch from Fountain Creek, extending it to Parsonsville Flat, opposite Lillooet, and Horsebeef Bar (ARMM 1876 p. 422). The ditch was completed in 1879 at a reported cost of $331,500 and presumably fed a large sluicing operation on the flat.  According to the Annual Reports, most activity in the ‘70s and 80’s at Lilloet was “desultory work” of Indians and Chinese.  Kennedy (2009) believes on the basis of the Annual Reports that “in any given year some six hundred people... were mining between Lytton and the mouth of the Chilcotin River” (p. 44).  In 1883, the Annual Report mentions an increase  in desultory mining resulting from an influx of Chinese leaving the railway works.  In 1885, the Lillooet-Fraser River and Gold Fields Co. sank prospecting shafts on five separate hydraulic leases in the area but did not develop any of them.  Desultory mining continued in 1896 but was somewhat contracted from the previous years.  This contraction continued in 1889 when there was “an exodus of [four fifths of] the itinerant Chinese miners on the banks and bars of Fraser River” (ARMM 1889 p. 289).  In the ‘80s through the first decade of the 20thcentury, many hydraulic mining operations were developed near Lillooet.   In 1889 two companies purchased farms for the purpose of hydraulic mining. In 1890 The Lillooet Hydraulic Mining Company was working a hydraulic claim just downstream of the Fraser River Bridge at Lillooet.  The Annual Report for 1890 states that “under the Mineral Amendment Act, 1890, five leases for hydraulic mining have been granted during the year.” and that “on three of these active work has been done” while on the other two plans existed to bring water to the site” (ARMM 1890 p. 376).  In 1893 “A large number of leases for hydraulic mining, especially near Lillooet, has been granted during the year and applications for several more have been received” (ARMM 1893 p. 1067).  In 1898 the Lillooet Hydraulic Mining Company 30 developed another lease near Lillooet.  This lease was worked through at least 1905.  In 1887, a hydraulic lease was developed on the right bank of Fraser River at Fountain; this was worked until at least 1902.  A substantial part of the placer gold production from the Lillooet District came from the Bridge River, which  was prospected in 1859, when paying deposits were  found seventy miles above its mouth (Bancroft et al. 1887 p. 454). Bancroft et al. note that Chinese miners “formed a large portion of the influx to the new field and soon became the chief holders of claims” (1887 p. 454) and that many of the mines were dam and wing-dam operations.  Oliver (1862) mentions that Indians were making considerable earnings working on the Bridge River.  In the early sixties, most of the mining on the Bridge was carried out on the lower portion of the river “from the mouth up to the head of Deep Canyon” (ARMM 1896 p. 551).  The Annual Reports to the Minister of Mines begin mentioning activity on the Bridge River in 1883.  In 1884, a jump in gold production in the Lillooet District is attributed to activity on the Bridge River.  In 1885, production on the Bridge river was very limited because of high water,an indication that most of the mining was still on the bed and low banks of the river.  A reconnaissance survey of physical evidence of mining along the Bridge upstream to the Big Bend Hydraulic mine revealed extensive tailing on terraces above the river, confirming considerable historical activity.  In 1896, the Bridge River Gold Mining Co. developed a hydraulic mine at the Horseshoe Bend on Bridge River.  In 1900 the Bridge River and Lillooet Gold Mining Company had 14 leases on the Horseshoe Bend, with twenty-five men employed.  In 1901, the company continued developing infrastructure in preparation for hydraulic mining with 36 employees.  The company began hydraulicking in 1902.  In 1903 the scale of the operation is evident: a cut was washed out “700 feet in length by 250 feet in depth” (ARMM p. 188). In 1905 the company was operating with a force of five men and they widened the cut to 200 feet.  The erosion of this cut would have moved somewhat more than 1 million cubic meters of gravel.  There is some mention of other activity on the Bridge in the early nineties including four claims on the south fork of the river, and a hydraulic operation on Alexander (Tyaughton) Creek, a tributary of the Bridge.  Chinese miners discovered and developed surface working claims on Cayoosh Creek above Lillooet in 1886 (ARMM 1886 p. 207).  The creek was worked intensively through 1889 with wing dams, rocker boxes and, in some cases, flumes to divert the main flow from the stream bed.  Around $160,000 in gold was recovered from the creek during this period.  One  company attempted to develop a hydraulic mine with a drain tunnel over 500 feet long.  The mine was worked in 1893 and 1894 but they had very poor returns, taking out much less gold than was expected “owing to the difficulty of meeting with large boulders, which had to be blasted” (ARMM 1893 p. 1067).  Kennedy (personal communication, 2010) reports good field evidence for mining along the creek. 31 Grasslands: Pavilion to the Sheep Creek Bridge  The first permanent mining operations in this region came in the spring of 1859 after the road from Port Douglas  to Lillooet had been opened.  Early miners in this area worked primarily on the bars of the river even through “it was not rare to find places above high water which yielded better than those below it”  (Bancroft et al. 1887 p. 456). Indeed, there must have been substantial activity on the banks of the river. Water records from 1859 record a 15-mile ditch carrying eighty miners inches of water at Grinders Creek  (GR 224 box 21), from 1860 report a 1 mile ditch carrying 500 inches of water from Cardis Creek (GR 252 Vol 12 File 1), and from 1861 a 12 mile ditch carrying 500 inches of water from Big Bar Creek (GR 224 box 22). These quantities of water and ditch lenghts clearly indicate large-scale mining was occuring.   Early maps (eg. Conroy 1862, Epner 1862) show trails along the river and trading posts established at intervals through the grasslands. Initial routes to the Cariboo may have followed the river, but with the construction of the Cariboo Wagon Road, this portion of the river became quite remote with the nearest substantial settlements being Lillooet, Clinton, and Alexandria.  Again, there is little record of activity in this area between 1862 and the beginning of reports related to the area in the Annual Reports to the Minister of Mines. GR 224 box 22 includes many records of mining activity in the region between 1879 and the late eighties, including hydraulic mines at the mouth of the Chilcotin, Cardis/China Flat and Crows Bar .   This activity is probably included in the mentions of “desultory mining” along the Fraser In the Annual Report from Lillooet between 1879 and the beginning of separate reports from the Clinton division in 1901.   Kennedy (2009) reports dredging operations at High Bar and Big Bar.  In 1899 the annual Reports first mention hydraulic mining at Big Bar with the combined seasonal production of nearly $3,500 worth of gold.  These leases continued operating until 1904.  By 1906 mining was “confined to a few itinerant Chinese and Indians” (ARMM 1906 p. 181).  A few sites, including Onion Bar, Crows Bar, and High Bar, were mined by mechanized operations into the mid twentieth century. Quesnel Forks and Alexandria: Sheep Creek Bridge to Cottonwood Canyon  Miners entered this region in 1859 (Bancroft et al. 1887) with discoveries concentrated on bars of the Fraser and mouth of the Quesnel River near the present-day town of Quesnel (then Quesnellemouth).  Substantial water development projects followed.  Water records from 1860-1861 associated with ditches supplying Rich and/or Ferguson’s Bar, Canada Bar,  British Bar, Snyder’s and Great Western Bar, Long Bar, and many other locations can be found in GR 216 vols 30-33.  Records from the region for the period 1867-1899 can be found in GR 216 vol 76 from the Quesnellemouth Division.  There is sparse documented activity in the period from 1867 to 1879. The eighties, on the other hand, were a period of substantial activity.  In 1880 the Annual Report to the Minister of Mines reports that “Very important discoveries were made during the season on the high benches along the Fraser and Quesnell rivers. Heretofore, mining operations on these streams have been entirely carried on by Chinese, who principally restricted their operations to rocking on the bars during low water. 32 In consequence of the recent discoveries of richer ground on the higher benches, quite an impetus has been given to this kind of mining in that portion of the district. Some fifty claims, including a half-dozen of white companies, have been recorded, and it is reported that some of the Chinese are taking out as much as $50 per day with a rocker.  Similar discoveries were made on Quesnel river, and I firmly believe that a thousand Chinamen and a good many white men will find profitable employment there during the coming season” (ARMM 1880, p. 429).  Initially, these bench claims were worked with rockers. Two to three meters of loess were shoveled off the surface of gravel deposits where the gold was found in the top four to eight inches (ARMM 1881 p. 394).  At some point between 1881 and 1891, these sites began to be worked more intensively.  The Annual Report for 1891 reports that “the few hydraulic claims keep working” (ARMM 1891 p. 563).  Hydraulic mining in the area continued into the beginning of the first decade of the twentieth century. Quesnel River: Quesnel and Cariboo Rivers Below Quesnel and Cariboo Lakes  The first discovery of gold on the Quesnel River was just above its mouth on Snyder’s and Macdonald’s Bar in 1859 (ARMM 1901 p. 969).  Within a year, over 600 white (in addition to, possibly, a substantial number of Native and/or Chinese) miners were working along the river, with especially rich diggings being found at Quesnell Forks, the confluence of the North (Cariboo) and South (Quesnel) forks of the river (Dawson 1889, Bancroft et al. 1887).  Water records indicate that some substantial mining development occurred low on the river in the early sixties.  Extensive ditches carried water from Dragon Lake down towards the Quesnel River (GR 216 vol. 30).  Oliver (1862) reports that 90% of claims along the river were making an ounce (~$16) a day to the hand. The North Fork was worked by over 50 miners during the sixties who earned from $10 to $25 per day, and was possibly the site of the first hydraulic mining in BC (Haggen 1923). The extent of mining along the Quesnel River is unclear during the late sixties and seventies.  In 1876, the Annual Report suggests that mining had not yet been extensively developed on the benches of the Quesnel River: “Nearly all the benches of the river have gold that will pay if water can be got on them, which will be accomplished in some way or other” (ARMM 1876 p. 420).  The Quesnel River contributed substantially to the total gold output for the Cariboo region in 1880 and 1881 (ARMM 1880 p. 426, 1881 p. 394) but large-scale hydraulic mines had not yet been developed (ARMM 1882 p. 357).  In 1887-88, a large hydraulic mine attempted to reach bedrock below Kangaroo Creek’s alluvium, cutting down over sixty feet into the gravel over which the creek was flowing (ARMM 1888 p. 294). In 1888, on the south fork, “there are two or three good claims, but owing to the scarcity of water for hydraulic purposes, the yield of gold is not very large” and on the lower river (below the forks) “the Chinamen still work away during the summer season, going from place to place according to the stage of water” (ARMM 1888 p. 294).  In the early nineties, it became clear that the thick gravel benches along the banks of the Quesnel River would pay well to hydraulic mining operations.  In 1889-1893, several hydraulic mines, two of them owned and operated by Chinese, were in operation on the south fork.  In 1891, one of the Chinese hydraulic mines on the south fork was bought by the 33 Cariboo Hydraulic Mining Co. This site eventually developed into the Bullion Pit, one of the largest placer mines in British Columbia’s history. The lower river continued to be mined in a desultory manner. This mining involved sluicing on benches until snow-fed water sources failed for the summer, and then working bars of the river (ARMM 1893 p. 1039).  In 1894, a large number of claims were located on both branches and below the forks, and on these claims, the Annual Reports state “work of development is being prosecuted with vigor” (ARMM 1894 p. 725).  By the next year there “are now about thirty leases of hydraulic ground... where two years ago there were but five.” Water is being brought into the Bullion Pit mine by eighteen miles of ditches which were capable of carrying 5000 miners inches of water.  Three monitors are operating to wash a gravel bank 280 feet high (ARMM 1895 p. 656). Substantial activity continued on many of these hydraulic mines, and other mines hydraulic mines were developed through 1904, when the amount of organized mining along the river began to decline, although some hydraulic mining, including work at the Bullion Pit, continued as late as World War II (Eyles and Kocsis 1989).  The Golden River Quesnel Co. attempted to mine the bed of the South Fork by de-watering the river with a dam s constructed in 1896 across the outlet of Quesnel Lake.  This dam operated until 1902.  M. Church reports observing mechanised work on Quesnel River bars in the 1970s (personal communication, 2011). Quantitative Information Regarding the Geomorphic Impact of Placer Mining  Though this thesis focuses on observation of the physical remains of placer mining and uses observation of these remains to directly estimate the geomorphic impact of placer mining , it is possible to use the historical account of placer mining along Fraser River and its tributaries to produce estimates of the geomophic impact of mining.  Field measurement of the impact of placer mining is costly.  If the historical proxy-based estimate of the geomorphic impact of placer mining is comparable to the estimate based on field observation it may be possible to apply a similar methodology to study the impact of placer mining in other regions through historical methods.  Historical documents include several sources of quantitative information regarding the scale, and potentially, the geomorphic impact of mining activity.  Some of the quantitative accounts that summarize activity include reports of the number of people actively engaged in mining and the methods that were employed, the total value of gold extracted, and the amount of water used at mines. It is possible to use these values as proxies for the total volume of gravel affected by employing a set of conversion factors that are based on the economics and practical constraints of mining. Figure 2.2 is a schematic of the relation between these proxies and several economic and practical constraints on mining that could be used to convert them to volumes of gravel.  All of the variables presented in figure 2.2 depend on the kind of mining that was done.  Table 2.2 presents estimates for some of the variables for different kinds of mining.  Values presented in table 2.2 have been derived from 34 the preceding narrative of the history of mining and from application of the mathematical relations conceptually shown in figure 2.2.   Ground sluicing is not included in the table because no specific mention of the gold concentrations found at groundsluice sites has been located. Presumably, groundsluice sites would have pay values somewhere between hydraulic and sluice sites.  The estimated concentration of gold is highly sensitive to the technology used to extract that gold: there is a three orders of magnitude difference between the minimum payable gold concentration for hydraulic mining (0.003 oz/m3) to the maximum concentration for high paying rocker box work (6.7 oz/m3).  The record of gold production provides the most temporally comprehensive and spatially well resolved quantitative summary of gold mining activity.  Estimates of the total province wide gold production were made from 1858 to 1873 and are shown in a table appended to the Annual Report for 1875. It is possible to use qualitative knowledge of the relative intensity of activity in different localities of mining for the period from 1858-1860 to distribute the total product by region. volume of gravel processed number of miners amount of gravel one miner can process duration of mining concentration of gold amount of gold wages amount of water duty of water Figure 2.2: Interrelation of proxies for placer mining gravel excavation volumes and practical and economic constraints. Rocker Box Sluice Hydraulic wages (dollars per day) --$3 (minimum)-$10 (good)-$50 (exceptional)-- highly dependent on conditions $6-$20-$100 value of gold --$10-$18 per ounce ($15 typical)-- 0.4-1.3-6.7 0.02-0.07-0.35 0.003-0.009-0.09 4.6-15-76 0.18-0.61-3.1 0.02-0.08-0.76 10 0.6 0.08 amount of gravel one miner can process (yd3/day) ½ yd3 10-15 yd3 gold concentration (dollars/yd3) 24¢-80¢-$4 3¢-$1 (10¢ normal) gold concentration (oz/m3) gold concentration (dollars/m3) generalized gold concentration (dollars/m3) Table 2.2 Estimates of key economic and practical constraints on different methods of placer mining used to estimate approximate gold concentrations necessary for each method to be remunerative. 35 For the period after 1860, the mining was too complex for such a distribution to be reasonable.  In 1874 explicit regional reporting of gold production values begins.  This record is fairly continuous. The boundaries and  resolution of the reporting districts change somewhat, and periods of one or two years are occasionally omitted for a particular area.  In some cases, in order to produce as comprehensive and spatially well-resolved a record as possible, values for particular localities and years were estimated.  Where production values bracketing a missing year or years were available, the value for that period was estimated as the average of the surrounding years unless there was indication of a major stepwise change in activity in the region, such as a major known discovery or exodus of miners, in which case the value for the most similar adjacent year was used.  In some cases production values were not resolved by locality but the numbers of miners and/or claims operating in the different localities were reported.  In these cases, the regional value was apportioned proportionally to the number of miners or claims.  Sometimes this apportionment was further nuanced if the average wages were known.  Table 2.3 a and b show the reported gold production values and interpolated estimates.  Another source (Haggen 1923) gives summary values for the whole mining history up to that time in different parts of the Cariboo which are reported in Table 2.4.  It is clear that estimates of the total amount of sediment affected by the mining based on gold production values will be highly sensitive to estimates of the estimated proportions of different sorts of mining that were done as the estimated gold concentration necessary to cover the cost of extraction varied with the amount of labor and capital investment required to work (see table 2.2).   Table 2.5 presents a first order estimate of the proportion of gold produced by mining with rocker boxes, sluices, and hydraulic operations (ground sluicing being grouped with hydraulic) in four regions of interest.  These values were estimated based on the narrative descriptions of mining activity in each region presented in the previous section.  It is possible to combine the estimates of the proportion of gold produced by each activity, the estimates of total gold production, and the gold concentrations necessary for each technique to be profitable to create an estimate of the total amount of gravel moved by placer mining.   For example, $120,000 worth of gold was produced in the Lillooet district in 1886 (Table 2.3a). It is estimated that in the 1880s 25%, 35%, and 45% of gold was produced by rocker box activity, sluicing, and hydraulic mining, respectively (Table  2.5).  Generalized gold concentrations for each of these technologies are 0.1, 1.64, and 12.5 m3 sediment per dollar of gold, respectively (table 2.2).  Combining these figures yields a total estimated sediment movement of            $120,000 x (0.25 x 0.1 m3 $1.00-1  + 0. 35 x 1.64 m3 $1.00-1 + 0.45 x 12.5 m3 $1.00-1 )     (2.1) or 739,000 m3.  Because gold production values for the period 1861-1873 are not resolvable at a scale finer than aggregate for the whole province, it is not possible to produce estimates of the amount of gravel moved in the areas of interest during this time period.  Figure 2.3 shows the estimated total amount of gravel processed and the method of mining in several regions through time.  Based on 36 Year 1858 520,353 520,353 1859 1,615,070 538,744* 200,000 1860 2,228,543 557135** 995,082 1874 1,844,618 55,000 38,084 12,000 1875 40,716 1876 2,474,904 9,114 25,000 95,905 1877 1,786,648 12,000 32,455* 12,833 1878 1,275,204 14,000 32,455* 50,910 22,630 1879 1,290,058 14,000 39,910 162,700 1880 1,013,827 10,800 81,800 95,300 1881 1,046,737 8,400* 40,717 207,500 1882 954,085 6,000 64,200 100,860 11,650 1883 794,252 5,000 68,000 92,740 4,860 1884 736,165 15,000 107,934 101,200 4,860 1885 713,738 29,000 94,700 71,000 11,400 1886 903,651 25,000 120,000 72,800 7,000 1887 693,709 20,000 73,750 79,520 5,700 1888 616,731 14,500* 37,660 80,127 6,520 1889 588,923 14,500* 26,239 57,500 8,200 1890 494,435 9,000 41,455 54,150 8,300 1891 429,811 6,400* 52,506 41,300 6,750 1892 399,526 6,400* 39,763 44,500 4,500 1893 379,535 3,800 51,376 37,050 4,500 1894 405,516 30,267 46,700 2,000 1895 481,683 48,400 40,663 86,910 36,000 1896 544,026 65,108 33,665 197,050 1897 513,520 58,680 37,480 200,000 1898 643,346 60,840 42,614 214,860 1899 1,344,900 74,720 42,700 193,300 1900 1,278,724 57,542 36,905 510,000 1901 970,100 45,440 26,080 240,000 1902 1,073,140 47,000 27,440 160,000 1903 1,060,420 50,400 25,820 132,000 1904 1,115,300 31,200 34,500 150,000 1905 969,300 4,600 30,000 96,000 1906 948,400 5,000 16,800 39,600 1907 828,000 3,000 12,000 44,000 1908 647,000 3,000 13,200 30,000 1909 477,000 2,000 10,000 12,000 Province Wide Yale Division, Fraser River Lillooet District (excluding Cayoosh Creek) Quesnel/Keithley District & Northern Fraser in Cariboo Division Swift and Cottonwood Rivers 876,325‡ 676,325‡ 10,000‡ 40,000‡ 30000† 48300† 3,800‡ * not reported, estimated as average of bracketing years ** estimated by number of miners and wages assuming a 3 month mining season estimate based on sum, then apportioned by number of people working or by number of claims operating based on qualitative descriptions of amount of activity in various regions † ‡ Table 2.3a: Gold production values for the Province of BC and several regions along and tributary to Fraser River.  These values are as reported by regional gold commissioners in the ARMM except where otherwise noted. 37 * not reported, estimated as average of bracketing years ** estimated by number of miners and wages assuming a 3 month mining season estimate based on sum, then apportioned by number of people working or by number of claims operating based on qualitative descriptions of amount of activity in various regions (Note: Quesnellemouth General includes the lower portions of the Quesnel and Cottonwood Rivers and Fraser River Noth of Quesnel. All Quesnel/Keithley Division includes Keithley Creek, which is not a sediment tributary of the fraser.) † ‡ year 1858 1859 1860 1874 4,603** 33,480** 1875 40,716* 1876 1877 1878 15,000 1879 62,700 100,000 1880 8,391 19,616 32,693 11,000 8,200 10,400 5,000 1881 82,300 7,945* 125,200 1882 7,500 51,660 8,600 6,000 19,100 8,000 1883 5,000 49,640 5,400 8,400 16,000 8,300 1884 6,500 54,600 5,400 8,000 15,000 9,000 2,700 1885 7,000 37,500 5,000 8,000 9,000 4,500 1886 10,700 32,000 3,500 6,000 8,600 12,000 1887 16,120 38,500 4,900 5,000 5,000 10,000 1888 17,000 39,125 5,402 6,000 2,600 10,000 1889 17,000 17,300 5,200 5,000 3,000 10,000 1890 17,500 12,600 5,500 3,750 3,000 11,200 600 1891 12,000 7,500 5,000 4,000 2,500 10,000 300 1892 10,000 2,500 7,000 5,500 8,500 9,500 1,500 1893 7,500 6,800 4,000 6,500 3,750 8,500 1894 7,000 16,000 6,000 4,000 3,500 9,000 1,200 1895 6,000 3,250 7,360 3,000 3,000 63,500 800 1896 197,050 1897 200,000 1898 214,860 1899 193,300 1900 510,000 1901 240,000 1902 160,000 1903 132,000 1904 150,000 1905 96,000 1906 39,600 1907 44,000 1908 30,000 1909 12,000 Quesnellemouth Generall Fraser below Quesnel above Quesnel all Quesnel/ Keithley Divison Lower Upper North Fork Quesnel [Cariboo] South Fork Quesnel Quesnel Tribs. 200,000‡ 995,081‡ 47952† 4930† 19719† 13446 † 9860 † 12833† 35910† 74,354‡ Main Stem of Quesnel Table 2.3b: Gold production values for regions in the Cariboo.  Sources are the same as for table 2.2a. 38 these figures, over the period 1859-1909 excluding the years 1860-74, mining excavated an estimated 6.8 million m3 of gravel along the Fraser and Thompson’s Rivers in the Yale District, 10.8  million m3 of gravel in the Lillooet District, and 36.7 million m3 of gravel in the Cariboo along the Fraser and Quesnel Rivers, of which at least 3.5 million m3 came from directly along the Fraser during a period of more detailed reporting from 1881 to 1890.  The figures for the Quesnel River are much lower than those calculated based on Haggen (1923) partly because some mining did continue past 1909, especially at the Bullion Pit, but also because different methods were presumably used to determine the amount of gold that had been extracted and because a larger part of the gold product was explicitly known to have come from hydraulic mining.  It is unclear whether the gold production values from the Annual Reports or from Haggen (1923) better represent the actual production.  The % hydraulic % sluice Fraser River at Quesnel 1,000,000 50 20 30 6,602,581 1,000,000 20 20 60 2,882,581 Sundry Claims, N. Fork Quesnel 1,000,000 60 20 20 7,842,581 Spanish Creek 500,000 100 0 0 6,250,000 222,648 0 0 100 22,265 South Fork -Sundry Claims 1,000,000 60 20 20 7,842,581 Roses Gulch 80,000 100 0 0 1,000,000 Chinese Farm 70,000 100 0 0 875,000 Chinese Pit, Bullion 900,000 100 0 0 11,250,000 Consolidated Cariboo (Bullion) 1,214,128 100 0 0 15,176,600 1,250,000 100 0 0 15,625,000 Quesnel River, Sundry Claims 3,000,000 60 20 20 23,527,742 Lightning Creek 8,000,000 50 30 20 54,030,968 Tributaries of Willow River 8,000,000 50 30 20 54,030,968 Regional Summary Values Quesnel River and Tributaries 9,236,776 76 11 13 89,411,768 Fraser River in Cariboo 2,000,000 35 20 45 9,485,161 Estimated Gold Production (dollars) % rocker box Estimated Gravel Moved (m3) Fraser River, sundry bars in Cariboo Golden River Quesnell Quesnel River Campan Creek Table 2.4 Estimates of gold production from Haggen (1923) and corresponding estimates of gravel excavation based on the estimates of proportions gold extracted by different mining methods and the method of applying gold production values as a proxy that is outlined in the text. RB H RB H RB H RB H RB H 1858-59 50 50 0 60 40 0 70 30 0 ------------ 70 30 0 1860's 40 55 5 40 55 5 50 50 0 40 40 20 45 40 15 1870's 40 55 5 40 55 5 50 50 0 40 40 20 45 40 15 1880's 15 25 60 25 35 40 20 20 60 30 30 40 25 30 45 1890's 5 5 90 15 25 60 10 10 80 5 5 90 10 5 85 1900's 50 50 0 35 55 10 10 10 80 5 5 90 10 5 85 Fraser River in Yale District Fraser River in Lillooet District Fraser River near Quesnel Quesnel River and tributaries Quesnel District (Fraser, Quesnel and tributaries) Sl Sl Sl Sl Sl Table 2.5: Estimates of the porportion of gold extracted in different regions and time perods by rocker boxes, sluicing, and hydraulic mining. 39 0 1,000,000 2,000,000 3,000,000 4,000,000 5,000,000 6,000,000 Color Indicates the Region of the estimated gravel contribution: Fraser and Thompson Rivers in the Yale Division Lillooet District (Fraser River from Fosters Bar to the Chilcotin, Quesnel District including Fraser River from Soda Creek to which for the time period from 1882-1890 is subdivided into Quesnell River and tributaries Fraser River from Soda Creek to Cottonwood Canyon Pattern indicates the type of mining: Rocker Box Sluice Hydraulic (and some groundsluice) Cottonwood Canyon, and Quesnel River and tributaries, Bridge River, and Cayoosh Creek) E s ti m a te d G ra v e l C o n tr ib u ti o n (m ) 3 1860 1880 1890 1900 N o D a t a f r o m 1 8 6 0 - 1 8 7 4 1858 1874 Figure 2.3: Estimated total amount of gravel processed and the method of mining in several regions through time. 40 annual reports better meet the tests of propinquity and contemporaneity but Haggen (1923) presents data with higher specificity. Conclusion  Comparing the narrative account of the history of gold mining along the Fraser and Quesnel Rivers with an attempted quantitative assessment of the scale of activity through space and time underscores the way in which the geomorphic impact of the “gold rush” and relatively small scale mining done by a very large number of individuals was dwarfed by the effects of the half-century of industrial gold extraction along the rivers.  Large-scale hydraulic operations occurred in most regions in the 80’s, 90’s and first decade of the 20th century and processed very large amounts of gravel. It is important to note that in some places where the Annual Reports write off mining activity as “desultory” other records indicate that large scale development was occurring.  This is particularly a problem  in regions that were quite remote, such as the upper Bridge River and the Fraser in the vicinity of Big Bar to the Chilcotin confluence.  Furthermore, because there are very few records of mining in the period from 1860-1874, it is difficult to understand how quickly the first phase of mining associated with the initial rushes attenuated.  Unrecorded but substantial mining work may have occurred during this time period.  The values presented in table 2.4 and figure 2.3 provide a first approximation of the geomorphic impact of placer mining in the region, but substantial uncertainty in estimation of the relative proportions of different mining activities and interpolation of the input data strongly limit the utility of the analysis for the purpose of understanding ongoing consequences of the mining.  Clearly, a very large amount of sediment was added to the Fraser and some tributary rivers by placer mining. Succeding chapters will present observations of the physical landscape left by placer mining that are an independent estimate of the volume of gravel excavated by placer mining.  Although the quantitative aspect of a historical account of the mining history is limited, the historical account does provide a helpful focus to define the boundaries for further physical investigation of the mining history.  Along the main stem of the Fraser large-scale placer mining occurred between Hope. and the Cottonwood Canyon.  Clearly, very substantial amounts of activity occurred on the Cottonwood and Quesnel rivers and their tributaries.  Although the first mining was done along the Thompson, its importance rapidly declined. Endnotes: 1In this account “bar” refers to active bars in the channel of the river.  Typically, these bars are locations that were gold bearing and at some point worked by placer miners.  Bars were worked by small scale traditional methods as well as sluices. 2Gold values are typically given in terms of a dollar value.  The value of gold was variable depending, presumably, on its purity.  Reported values range from $10 to $18 dollars an ounce, with with values of $15 to $16.50 representing the vast majority of the gold recovered in the areas of interest 41 (see for example the tables “Province of British Columbia Mining Statistics” that are appended to the Annual Reports to the Minister of Mines during the 1880s.  This account retains the usage of dollar values to report the amount of gold recovered.  Of course, the dollar values reported are contemporary values. 3Generously assuming an average mineable bar width of  60 m along this 26 km stretch of river yields a total minable surface area only large enough to contain ~500 100 by 100 foot square claims. 4For purposes of comparison, a miner using a gold pan “can pan about one-half a cubic yard of material a day” (British Columbia Department of Mines 1946 p. 26). 5The authors of the Annual Reports use the term “desultory” to refer to mining that the authors perceived to be unorganized.  Such operations were probably typically small scale and uncapitalized, and may have been fairly mobile taking advantage of changing water levels in the river and changing availability of water supply for sluicing dry diggings.  Physical evidence (see Chapters 3 and 6) indicates that some such operations may have actually been quite important but discounted by the authors of the Annual reports due to the ethnicity of the company working the mine or remoteness of the site. Because of the ambiguity in the meaning of the term, its usage is retained in this account rather than the application of a more specific term such as “small scale mining” or “rocker box work on bars.” 6A miner’s inch is a measure of discharge equivalent to approximately 1.5 cubic feet (0.042 cubic meters) per minute.  The value is somewhat variable because a standard method of measurement was lacking.  Kent (1895) states that the value of a miners inch ranges from 1.38 to 1.78 cubic feet per minute. The value of 1.5 cubic feet per minute became the standard in most places (Ricketts. 1943).  In British Columbia, the legal definition of a miners inch in 1915 was  “the flow through an orifice 2 inches high by ½ inch wide made in a 2 inch plank, the head on the top of the opening being 7 inches ... 35.5 miners inches are equivalent to 1 cubic foot per second” (Etcheverry 1915, p. 71). 42 CHAPTER 3: MAPPING AND MEASURING MINES  Historical research provides a remarkable amount of context and background information on the placer mining history along Fraser River.  It provides information on the names and general locations of mine sites, mining techniques employed, and times of activity at various sites in varying levels of detail for different regions and time periods.  It provides very little direct quantitative information on environmental impacts, however, and the accuarcy of extrapolated estimates is not clear without corroborating evidence.   Many historical geographers have established the importance of corroborating evidence gleaned from historical sources before making inferences about environmental history (eg. Butlin 1993, Gurnell et al. 2003, Hooke and Cain 1982).   Direct field observation of mine sites provides an opportunity for corroboration of the historical mapping and  derivation of quantitative information about the narrative constructed from historical research. In many –perhaps most– cases the landscape effects of the mining persist at the mine site and are still observable.  Using these landscape signatures as a guide, it is possible to map the planimetric extent of placer mines, from which the mine site area may be directly derived.  It is also frequently possible (see chapter 4) to map the three-dimensional geometry of the mines to estimate the total volume of removed material.  A comprehensive (or as nearly comprehensive as possible) map of mines along the Fraser can provide the basis for a regression-based estimate of the total sediment contribution to the river from the mining.  Maps of mine locations and shapes are also of interest from a social-historical perspective because mine sites are a durable industrial-archaeological record of the mining activity.  They are locations that may be of interest for historical preservation and/or development for tourism industries; and they provide a starting point for searches for other mining-era archaeological sites because remains of settlements and gardens are frequently close to the historically worked mines.  Kennedy (2009) has thoroughly mapped placer mines from Lytton to the Big Bar reaction ferry crossing.  Work associated with this thesis extended the mapping north to the town of Quesnel and south to the town of Hope. Methods for mapping physical evidence of mining  Mines were identified using the landscape effects of the mining as described by Kennedy (2009).  Kennedy (2009) describes the the landscape effects of several different mining techniques: gold panning directly on the bars and banks of the river left negligible landscape effects.  The use of rocker boxes leaves small piles of tailings and isolated boulders.  The long tom and sluice box (very similar techniques that will hereafter be referred to simply as sluices) leave tailings in linear heaps and gravel lags, which are often in geometric rows.  Ditches and flume remnants lead to sluicing locations.  Groundsluicing leaves large, frequently irregular heaps of lag cobbles which are sometimes behind walls of stacked cobbles; they form eroded depressions with steep scarps, wide floors that slope toward a drain, and are also always associated with ditches.  They lack the headboxes associated with hydraulic mining.   Low head hydraulic mining leaves a landscape 43 signature similar to ground sluicing.  High head hydraulic mining typically leaves very high scarps and sloping wash pits that are frequently covered in cobbles to large boulders. Table 3.1 lists diagnostic and indicative landscape characteristics for each of the major categories of mines.  Gold panning and the use of rocker boxes and sluices usually occurred below the high water line on the bars and banks of the river.  The evidence of mining at these locations has usually been obliterated by sediment movement at high water.  Though these activities may have destabilized armored surfaces, they did not result in a net addition of sediment to the river and, therefore, probably did not have major long-term environmental consequences.   When material was removed from terraces and slopes above the river, however, large amounts of sediment were typically dumped into the river.  From sluicing to high head hydraulic mining, the methods of mobilizing sediment are similar to natural erosional processes and therefore occasionally one encounters erosional Table 3.1: Landscape characteristics indicative of historical mining sites. Diagnostic Supportive Necessary Sluice Linear heaps of tailings Hydraulic All Manually stacked cobbles and boulders Groundsluice Manually stacked cobbles and boulders Gully-like erosional pattern in locations where water would not be naturally concentrated Ditches leading directly to the scarp edge Remains of headboxes, high pressure iron pipe, hydraulic monitors Teardrop shaped depressions at terrace edges with high scarps Water source with high flow and substantial head (may be quite distant) Presence of original terrace level “buttes” or barriers between an eroded pit and the downhill direction Ditches leading to or above the area Plausibility of placer gold presence (unless the site is just of a prospect size) “Unnatural” scarp patterns including sharp angles, freshness in areas where other erosion has not recently occurred, odd angles relative to nearby water features Clearly constructed drains leading through barriers Nearby mining-era artifacts of human activity Plausibility of access to a reasonable quantity (defined by the scale of the site and technologies used) of water Remains of flumes, sluices Soil and fines removed over an area leaving unorganized coarse (cobble-boulder) lag deposits 44 features that may have been created by mining, by other anthropogenic causes such as escaped flood-irrigation, or by natural gully erosion or mass-failure.  Some of the ambiguities between natural processes and features that are diagnostic of mining activity can be seen in the “supportive” landscape features shown in table 3.1.  Positive identification of mining activity was made if any of the diagnostic features was present.  If only supportive features were apparent at a mine location, mining activity was mapped with a lower degree of certainty.  A scale from one to three was used to represent these degrees of certainty. Class 1 represents a site where there are diagnostic features of mining and where the boundaries of the mine are clear from ground-based field work and/or air photo interpretation.  Class 2 represents locations where there are diagnostic landscape features but the boundaries of the mine are not perfectly clear and locations where a site with clear boundaries has been identified only on the basis of supportive features.  Class 3 represents sites where there is substiantial uncertainty regarding the locations of site boundaries or where indicative features are relatively week.  Once a mine has been positively identified, classification of the type of technology used is of interest.  This is both because it is a step toward interpreting the industrial-archaeological record and because the different technologies left potentially distinctive morphologies that may have distinct area-volume relations.  Where historical records document a specific kind of technology used at a location, the mine was assigned to that technology.  Two factors complicate attempts to classify mine sites in the field by the technology used to work them.  First, the techniques from sluicing to high head hydraulic mining grade into each-other.  A “sluice” operation is a location where sediment is excavated primarily by hand and placed into a sluice box where water is flowing.  A “groundsluice” mobilizes the sediment with flowing water (but almost assuredly manual tools are involved in most cases as well) and may pass the sediment  through a sluice box or may use the bottom of the wash pit as a natural sluice box to concentrate gold.  A low head hydraulic operation is essentially identical to a groundsluice except water at the site of the mine could be precisely targeted and locally moved over the surface of the mine without construction of ditches.  The amount of head used in a hydraulic operation is gradational until, at some critical threshold, it is referred to as “high head.”   Sites that have no distinctive features of groundsluice or hydraulic activity were mapped as sluice sites, even if they did not have features diagnostic of sluicing.   A second factor that complicates attempts to classify mines by the technology used at the site is the fact that the histories of individual mine sites are frequently multi-layered.  Sluice sites are sometimes over-printed by hydraulic operations, which themselves may have been worked over at a later date by rocker box1.  In locations where there was evidence of the application of multiple mining techniques, the mine was categorized as having been mined by the technique that affected the majority of the surface area.  In some cases (visible on the map where individually mapped sites share edges) the decision was made to separate portions of what could be considered a single site at locations where different technologies were used in order to preserve reasonable area-volume relations. Because of the difficulties associated with determining the technology used at a mine site, the technology attribute of the mapping should be considered a 45 functional classification describing the general morphology of the site more than a direct inference about the type of activity that occurred historically.  Before a site can be established as having had a mining history, it must be located as a plausible location in the first place.  Reconnaissance to determine probable mine locations was done using a variety of techniques including air photo interpretation, interpretation of historical documentation combined with guesses about locations where placer gold accumulations are probable, a reconnaissance float trip down the river, and views from ridges and bluffs along the river.  A variety of air photos was used.  In areas where forest succession has screened the ground surface from aerial views and in places where industrial activity has recently altered the landscape, the oldest available photos were extremely useful. From Hope to a little above Spuzzum, a set of 1928, 1:8000 scale photos were used (eg. National Air Photo Library, 1928, A289 40-90).  Most of the river was covered by mid-twentieth century flight lines with a typical scale of ~1:13,000 (eg Base Mapping and Geomatic Services Branch 1951, S84-R4 305-316).  In other cases, particularily in the grasslands where tree cover has not obscured mines, recent, higher quality and color photos were extremely helpful in identifying mine locations.  The scale in these photos is either 1:20,000 (Base Mapping and Geomatic Services Branch, 2004) or 1:30,000 (Base Mapping and Geomatic Services Branch 2005 and Base Mapping and Geomatic Services Branch 2006).  Satellite/air photo imagery included in Google Earth’s database and other imagery in the British Columbia Imagery WMS server (GeoBC Web Map Services 2011) was also consulted. A complete list of air photos that were consulted during mapping is presented in table 3.2.  In heavily forested areas, often the only way to find mine sites was to determine locations Agency Date Roll Photos Region Scale NAPL 1928 A289 40-90 Yale Area 1:8000 NAPL 1928 A297 60-97 Yale Area 1:8000 NAPL 1928 A298 1-2 Yale Area 1:8000 BC 1948 BC590/X170 1-11 Boston Bar Area 1:13,000 BC 1951 S69R4 1-24 Yale Area 1:13,000 BC 1951 S84R4 305-316 Yale Area 1:13,000 BC 1963 BC5070 163-185 Soda Creek-Cottonwood Canyon 1:40,000 BC 1949 BC856 78-96 Sheep Creek-Soda Creek 1:40,000 BC 1951 S84-R1 720-730 1:13,000 BC 1951 S84-R3 356-389 N Bend to Lytton 1:13,000 BC 1952 S82-R1 665-680 Boston Bar Area 1:13,000 BC 1963 BC5071 11-15 Soda Creek-Cottonwood Canyon 1:35,000 BC 2004 Lytton-Big Bar 1:20,000 BC 2005 Big Bar-Alexandria Area 1:30,000 BC 2006 Alexandria Area-Quesnel 1:30,000 N Bend to Lytton various (~100 rolls, see list of works cited for complete list) (NAPL is the National Air Photo Library and BC is the British Columbia Base Mapping and Geomatic Services Branch.) Table 3.2: Air Photos Consulted During Mapping. NAPL is the National Air Photo Library and BC is the Base Mapping and Geomatic Services Branch. 46 based on references in the historical record.  Map 2.1 and accompanying archival records (see Chapter 2) were used as the basis for locating mines from historical evidence.  Where the historical record had a strong indication of important activity –such as a large cluster of claims or water rights records, claims or water rights that were renewed for consecutive years, and narrative description in the Annual Reports to the Minister of Mines of substantial activity– probable locations for the described activity were determined based on the landscape as described by topographic maps and air photos.  In September 2009, a float trip was made down the Fraser from Quesnel to Chilliwack. During the trip evidences of mining activity were recorded.  Often, mines on terraces near the river banks were visible enough to positively identify mines based on the criteria set out in Table 3.1 and boundaries were distinguishable in aerial photographs.  In other cases, some evidence of mining activity –cobble armored banks, apparent tailings piles leading from drains, and ditch lines leading into areas– was visible from the river, but the mine itself was not.  During field work in the summer of 2009 and 2010, some mines were identified from vantage points on the ground along roads that parallel the river.  Binoculars and a spotting scope assisted views of mines.  As with the river transect and air photo interpretation, some mines could be positively identified and mapped from this view, while others needed ground site inspections.  Where mines boundaries were identifiable in air photos, they were sketched onto mylar photo overlays while viewing them in the field.  These sketches were then digitized in ArcMap 9.3 over the top of orthorectified air photos or the WMS air photo imagery layer (GeoBC Web Map Services 2011).  In places where mine boundaries were not discernible in air photos but were visited on the ground, a Garmin GPSmap 76CSx GPS unit, which typically had a positional accuracy of ± 7 to 10 m, was used to define corners of the sites.  For small sites where a 7 or 10 m accuracy was comparable to the whole size of the site, the GPS unit was used in combination with edge distances determined on the ground by transect tape and/or pacing.  Data from the GPS combined with notes from the pacing were used to digitize approximate polygons of sites that were not visible in air photos.  Areas of mapped sites were determined using the “Calculate Geometry” tool in ArcMap 9.3. Geographic Variations in Methods  Methods used to find mine locations varied regionally along the river as a consequence of variable vegetation cover and topography.  As the river flows from Quesnel to the coast it passes through six Biogeoclimatic (BGC) zones.  At Quesnel, the river flows through the Spruce Sub Boreal zone.  This merges into the Interior Douglas Fir zone near Alexandria. The Bunchgrass zone begins at the Highway 20 (Sheep Creek) Bridge and extends south to the Fountain Bend just north of Lillooet. From there to just south of Lytton, the river flows through the Ponderosa Pine zone. Next the river again flows through the Interior Douglass Fir zone which extends as far south as the site of the Alexandria Bridge (Highway 1).  From here to Hope the river flows through very dense forest in the Coastal Western Hemlock  (CWH) zone. 47  Similarly, the topography of the valley around the river is quite variable along the reach of interest.  From Quesnel to Soda Creek, the river flows through a broad sediment-filled valley. At Soda Creek the river enters into a series of deep valleys and bedrock canyons that extends as far south as Yale. At Yale, the valley begins to widen, becoming fully alluvial at Hope.   Each combination of topography and BGC zone created a unique working environment, a unique successional pattern of vegetation recovery on mine sites, and distinct pattern of placer gold deposits and accompanying mining history and technique.  These regional variations required adaptable techniques for air photo and ground-based identification.  In some areas ground-based investigation was necessary to identify and map mines.  Some substantial regions were not visited as a part of the project because of access difficulties and time constraints.  In particular, no field visits have been made in the following areas: on the left bank (SE side) of the river between Trafalgar Flat and the Alexandria Bridge, in the canyon between the Alexandria Bridge and Boston Bar, and, with the exception of a single site identified at Kanaka, from just north of North Bend to Edinburgh Flat (just south of Lytton).  Locating mine sites along the stretch of river from Hope to Lytton is quite difficult.  Here the river runs through the CWH and Interior Douglas Fir BGC zones.   Forests are dense and so it is nearly impossible to identify mines without a ground-based visit.  Ground-based visits are difficult because of heavily forested terrain, a prevalence of small hold properties, and limited road access combined with very difficult cross country travel in the steep canyon.  Because of rapid soil development and thick moss cover even ground-based visits may miss important mine sites. Historical sources were key in directing efforts to locate mines on the ground. Air photos from 1928 (see Table 3.2) allowed the identification of some mines from Hope to just above the Old Alexandria Bridge.  In these photos some mines had not yet been re-vegetated, some had been recently uncovered by logging, and some were visible as areas where trees had not been removed or ground had not been cleared for agriculture.  In addition to historical sources, patterns of cobble armoring on the banks below terraces that were visible from the river proved to be useful indicators of the presence of an upslope mine.  With the exception of a couple of locations that were fortuitously stumbled upon, all of the mines mapped along this part of the river were initially located by one of these means.   Because of the difficulties associated with dense forest and the limited ground-based field work in this region, some mines probably remain unmapped.  From Lytton to Big Bar, mapping very closely followed Kennedy (2009).  His mapping was modified  only when ground visits indicated substantial changes to the mapped polygons should be made.  In the original mapping, Kennedy used mylar overlays on a set of 1:13,000 black and white air photographs that were taken in 1950 to outline mines that were observed from overlooks or visited on the ground.  Along this portion of the river nearly all of the mined sited have probably been identified.   This area is in the Ponderosa Pine and Bunchgrass BGC zones.  In the Ponderosa Pine zone, trees are typically established on mines but little or no soil development has occurred.  Grasses and brush are sparse or absent and cobble lags are clearly visible. 48  From Big Bar to the Sheep Creek Bridge mapping was primarily done by identifying likely mines in air photos and then confirming them during the raft transect or by hiking to vantage points along the valley with a spotting scope and binoculars.  Because this portion of the river runs through the Bunchgrass BGC zone mines were clearly visible from distant vantage points.  Curiously, in this area succession on many mines has involved the establishment of tree cover even though the surrounding terrain has very few trees.  Because of this, mines could be identified in air photos as locations where dense stands of trees existed with geometries that match mine site geometries or where mine-like scarps were visible in the grass.  Nearly all of the mines in this region have probably been identified.  From the Sheep Creek Bridge to Cottonwood Canyon, just north of Quesnel, mapping was substantially more difficult.  Here the river runs through the Interior Douglas Fir and Sub Boreal Spruce BGC zones.  Here again, dense forests obscured visibility of mines.  Mines were located by consulting historical sources. Also, frequently, when asked for permission to visit a potential site on their land, individuals would describe locations of mines on their land and that of their neighbors. Succession in this area somewhat matched that of the Ponderosa Pine zone.  Trees and brush were established over mine sites but soil development was quite limited.  Because of this, it was sometime possible to identify mine sites in air photos based on high reflectivity of gravels that are visible in gaps between trees.  As with the section of the river between Hope and Lytton, many mines in this area nearly certainly remain unmapped. Summary and Evaluation of Physical Evidence of Mining  Polygons of the mapped mine sites are included in Map 2.1.  A total of 456 distinct placer mining excavations has been identified along the Fraser between the town of Hope and the Cottonwood Canyon. Some of these are clusters of small excavations that may best be considered one “site” but most of them represent substantial, independent sites.  Identified mines tend to be located fairly close to the river (on the first or second terrace), but may be on terraces high above the river. Many smaller (below ~1000 m2) prospects were noticed but remain unmapped. Furthermore, many areas that were almost assuredly worked by rocker boxes along the banks of the river have not been mapped as mines because there is no distinct and extant evidence of the activity.  Additionally, some substantial mines in forested regions have probably not been found and are therefore not mapped.  Of the mapped mines, the average area is 26,000 m2, the median area is 29,000, the largest is 564,937 m2, the smallest is 114 m2, and the standard deviation of area is 48,545 m2. Mine areas are strongly positively skewed and approximate a lognormal distribution.  Locations of mines relative to the river through the whole corridor seem consistent with the observations of Kennedy (2009).   With few exceptions, pan, rocker box, and sluice sites below the high water mark are not preserved.  Preserved sluice sites occur on the upper banks of the river, on valley slopes, and on terrace surfaces.  Additionally, many sites were identified on alluvial fans of creeks that empty into the Fraser.  Ground sluice sites typically cut into terraces from the terrace 49 scarp backwards.  Many groundsluice and sluice sites seem to have targeted the intersection of bedrock, debris flow or colluvial deposits with alluvial depots on valley slopes and alluvial fans. Presumably, cobbles and boulders on the debris flow surfaces resulted in armoured surfaces that were stable for substantial time periods.  High roughness associated with large grains on these surfaces may have caused turbulence and served as trap locations for placer gold.   Hydraulic sites cut deeply into terraces. They are typically located where there are thick deposits of alluvial gravel but may also be located where alluvial gravel (that must have had very high pay values) is overlain by thick non alluvial deposits.  There are several clusters of landscape evidence of mining activity.  These clusters are presumably locations of relatively intense historical activity.  The first such cluster is between Hope and just north of Yale at the break in river gradient where the Fraser exits the bedrock canyon and begins flowing over its own alluvium.  The second is very tightly grouped in the area where the Fraser Canyon widens for a short distance in the area of Boston Bar and North Bend.  The third is between Lytton and the Fountain Bend just north of Lillooet.  There is a sub cluster of very intense activity at Lillooet and the Fountain Bend.  A fourth cluster is centered at Big Bar and extends from High Bar in the south to Crows Bar in the north.  A fifth cluster extends from Canoe Creek to Iron Canyon just north of the Chilcotin confluence, but is most intense in the region of China Flat and the Chilcotin confluence.  Both the fourth and fifth clusters are located in the deep valley through which the Fraser flows in areas where it is wide enough for there to be substantial benches and terraces on either side of the river.  Finally, there is a cluster of activity stretching from Soda Creek to Cottonwood Canyon, with the most intense activity in the portion north of Kersley .  This stretch of the river is generally alluvial.  The valley is wide with large, gently sloping terraces bordering the river.  Also, this last cluster is just downstream of the mouths of the Quesnel and Cottonwood Rivers which drain much of the most productive parts of the Cariboo Mining District.  An extraordinary and enduring record of the mining activity along Fraser River is preserved in the landscape along the river.  Many mines remain undisturbed and mappable and some others are mappable on the basis of historical photos and air photos.  The picture provided by maps of this evidence shows that mining along the Fraser was very widespread and occurred extensively outside the regions typically cited as belonging to the “Fraser” or “Cariboo” Gold Rushes.   There are clusters of activity that appear to be controlled by local conditions related to the probability of gold-placer formation along the river.  The clustered nature of the physical evidence that indicates historical mining activity contrasts interestingly with the map of archival records of mining activity.  Though the archival records would indicate clustering of mining activity that essentially matches the pattern of observed mines from High Bar north, it does not reveal the clustered pattern in the canyon from Lytton down to Yale.  This incongruence may be a result of difficulty in finding mines in the forested region, or, probably more likely, is related to the fact that physical evidence of mining is typically only preserved where the mining is above the high water mark.  Much of the activity in this region occurred on the lower banks and bars of the river. 50  Determination of the technology used to mine was often possible based on distinctive morphologies.  These interpretations, however, should be treated with caution as, in many cases, they are best guesses as to the technology used at a site.  The map of mines and mine site areas provides a set of good estimates of mine areas.  Even mines mapped as Class 2 or 3 provide reasonable estimates of the order of magnitude of the mine area because the maximum size of the sites is often constrained by local topography.  The only place where there is a strong possibility of very unreasonable over-estimation of the mines area would be very large class 3 sites where there is good indication of local mining activity through a large area but where there may be many small, unconnected sites.  It is important to remember that, for the most part, with the exception of some class 3 sites, the mapped mines provide a minimum estimate of the observable mining activity.  In some regions more than others, many mines nearly certainly remain unmapped.  The potential errors associated with mine mapping  are difficult to quantify but most probably are very asymmetric with relatively few false positives but quite a few unnoticed sites.  Excavations of 456 placer mines have been mapped along Fraser River between the Cottonwood Canyon and Hope on the basis of physical evidence left in the landscape.  Though the map probably includes some falsely identified sites where natural erosion left a landscape very similar to placer mining, there are probably substantially more sites that remain unmapped than have been falsely mapped.  This mapping provides valuable base information for historical and archaeological investigation of the placer mining history along the river and a strong basis from which to quantify the environmental impact of the mining activity. Mine Surveys  A key factor determining the magnitude of the impact of placer mining on Fraser River is the volume of sediment excavated by the mines and released into the river.  No estimates have previously been made of the volume of sediment produced by mines along the Fraser and its tributaries. A century to 150 years after the mining, substantial physical evidence remains in the landscape.  In many cases both mine scars and the terraces into which they cut remain essentially unaltered since the mining.  This opens the possibility of determining the magnitude of the geomorphic impact of the mining through direct survey of the relict landscape so that comparison can be made to the estimates in Chapter 2 that were based on the proxy of estimated value of gold production and assumed concentrations of gold. It is possible, therefore, to produce two independent estimates of the total volume of sediment moved by placer mining along the Fraser: one based on physical evidence and the other based on historical documentary records.  This chapter outlines the approach used to estimate the volume based on physical evidence.  Placer site morphologies have already been described (Chapter 1 p. 3).  Mines are typically cut one to seven meters into geometrically simple terraces or terrace sequences and cover 500- 150,000 m2. At these sizes, manual surveys of individual sites were generally feasible and frequently 51 took one-half to two days to complete.  Because of the substantial labor involved in mine site surveys, it was not reasonable or desirable to survey all of the mapped sites. Rather, a regression relation predicting volume based on area was developed from a subset of surveyed sites to allow estimation of the volume that each mapped mine contributed. Site Selection  Mines were selected for detailed survey in order to represent the whole range of mapped mine areas.  Further criteria for selection included accessibility (permission of the land owner; distance to drivable roads; terrain to cross from a road) and knowledge of the mine’s existence. Recall that the landscape features left behind by three different mining technologies – sluice boxes, ground sluicing, and hydraulicking – are somewhat distinctive to each technology. Because of this, independent regressions predicting volume from area may need to be developed for each technology . Therefore, it was important to sample each mine type in a way that could yield reasonable inputs for an independent regression.  An error in the estimated volume of of any particular large site will have a greater impact than an error in the estimation of a small site.  However, because  the frequency-volume relationship of mines was unknown prior to fieldwork, field sampling did not emphasize large sites to the exclusion of small sites.  It is possible that researcher bias influenced the choice of survey sites in a way that could introduce a bias in the average depth of mine surveyed for any given area because it is much easier to survey deep mines than it is to survey shallow mines.  Most of the surveying was done in the region previously mapped by Kennedy (2009).  Because the sites had previously been mapped, it was possible to select sites  for survey based on accessibility, type, and size prior to  seeing them.  This procedure was adopted to limit researcher bias in mine site selection. Survey Methods  Several different methods were used to determine the present day surface at mine sites.  Most mine sites (33) were surveyed with a robotic total station.  The point density was variable between mine sites (~50-1000 survey points per mine) and sufficient points were recorded to generally characterize the shape of the surface. For sites with very simple surface topography, a relatively low sampling density was possible while for very complex sites a higher point density was necessary.   A regular grid was not used, but survey points were strategically located at places that would result in a realistic triangular irregular network (TIN) when they were connected. A manual backup for the total station was used once.  It consisted of a tape, level, and staff  that were used to construct regularly spaced cross sections across a site.  Six sites were partially surveyed to determine a local average depth that was then extrapolated across the whole site area to estimate a total volume.  From a visual assessment of these sites it was deemed that this provieds a reasonable approximation. 52  In densely vegetated areas, short lines of site required a very lightweight and mobile surveying setup.  A clinometer was mounted on a staff of known height and used in combination with a tape to produce cross sections, which were, again, spaced in such a way as to generally characterize the surface. This approach was used at 3 sites.   In some cases (15) where a mine’s geometry is simple, where much of a mine had been destroyed by post-mining development, where very little time was available for a survey, or where mines were very extensive in vegetated terrain such that a complete survey would have taken an inordinate amount of time, mine volumes were characterized by a measurement of the scarp height at various places using a level mounted on a pole of known height (sometimes called a Jacob’s Staff). This technique is commonly employed in stratigraphy to measure bed or unit thickness.  At very geometrically simple sites (2 surveys), a single measurement of the scarp height characterized an average depth.  At one site, post-mining industrial activity has obliterated most of  the original mine, which is clearly visible in historical air photos.  After interviews with long-term workers at the site (a concrete plant and lumber mill) it was determined that an extant scarp represents the original mining scarp. This depth, which is probably a minimum estimate, was extrapolated to the whole site. For the large and forested sites (13 surveys) two techniques were employed using the Jacob’s Staff to determine elevation differences.  In complex sites, regularly spaced grids were determined across the mine site using a grid over an air photo.  Navigation between grid points was done using a compass and pacing.   The depth of the mine at each grid point was estimated either by measuring a nearby remnant surface or by measuring the elevation difference from an already estimated point.   For sites where the mine floor was more uniform scarp heights were estimated around the perimeter of the site and extrapolated across the floor of the site.  The geometry of three  large mines was determined using published maps.  Two of these (the Horseshoe Bend and Bullion Pit) were determined using a 15 m DEM that has a vertical accuracy of +/- 5m (Centre for Topographic Information, 2000).  Because of the very large size of these two mines (average depths of 77 and 38 m, respectively) this DEM provided enough detail for a reasonable estimation of the volume. Both of these mines are slightly outside the study region but because they were known to be very large and the effort required to secure information about them was modest they were included in order to provide regional perspective on the impact of the mining history and in the hopes that they would fit, and thereby extend to larger sites, the regression relationship established for the study area.  The geometry of one mine was determined using the Lytton Moran Survey (BC Department of Lands 1956), which has a scale of 1:6000 and a contour interval of 40 feet.   This mine was also quite large, with an estimated average depth of  9.5 m (31 feet).  Surface Creation and Volume Estimation  Survey data were plotted and analyzed in ArcMap 9.3 GIS (ESRI 2008) utilizing the 3D analyst, Geostatistical Analyst and Spatial Analyst extensions and the TIN Editor Toolbar (ESRI 53 2010).  Surveys were georeferenced using a Garmin GPSmap 76CSx GPS unit to define the location of the survey origin (in the case of total station surveys), cross section origins (in the case of manual surveys), or point depth estimates where a sampling grid was not used.  The GPS reported absolute positional accuracy was typically +/- 7m and rarely less precise than +/- 11m. The differential precision is unknown but presumably less than the absolute precision. Survey points were then overlaid on digitized air photos to aid in interpretation of the survey results and reconstruction of the pre-mining surface.  For all sites that were surveyed either using the total station or one of the manual methods, the first processing step was to create a model of the current surface using the survey or cross section data.  After the survey data were filtered to remove spurious points, 3D Analyst was used to create a preliminary TIN of the surface. Because the mines often have strong linear features (eg scarps, drains, ridges), the automatically created TIN often connected nodes across these features resulting in an unrealistic representation of the surface.  The TIN Editor toolbar was used while viewing aerial and ground-based photos from field visits to define appropriate node connections to generate a realistic model.  The next step was to reconstruct a plausible pre-mining surface. To do this, all of the survey points from within the mine were removed.  In the simplest cases, a second TIN was created based solely on surveyed points surrounding the mine.  This was possible if the geometry of the surface into which the mine was cut was approximately planar and was preserved on at least three sides of the mine (assuming the mine is approximately a rectangle).  This procedure was done at eight sites. In cases where three sides of the mine were not preserved or the surface was slightly more complex, but could be approximated by a plane or other polynomial surface, polynomial interpolations were performed using the Geostatistical Analyst extension to project the remnant surface over the top of the mine. This procedure was done at twelve sites and resulted in a DEM of the pre-mining surface. In some cases these two procedures were inadequate to produce a plausible surface.  In these cases a TIN was created using all survey points from outside the mine and nodes were manually added to the TIN to represent the pre-mining surface. Frequently, high remnant surfaces within the mine or preserved remnant “buttes” that for some reason were not mined were used as guides.  The general form of the landscape and evidences within the mine (depth of lag deposits, pattern of stacked cobbles) also provided clues to the locations of terrace edges or other important features in the pre- mining landscape.  The original surface at sixteen sites was constructed using varying degrees of manual TIN manipulation.  The pre-mining surface TIN and the current surface TIN were both converted into raster- based digital elevation models (DEMs).  They were then clipped to the extent of the mine as determined by air photo interpretation and field survey notes.  Simple subtraction of each cell of the current surface DEM from the pre-mining surface DEM then yielded a DEM of the depth of removed material for each cell.  The average of this DEM was multiplied by the area of the mine to produce the total volume of the mine. 54  In locations were the depth of material removed was directly estimated either at grid points or along scarp edges, these estimates were placed into GIS and a polynomial interpolation was used to interpolate between the estimated points.  This surface was then clipped to the area of the mine. The average depth of this surface was multiplied by the area of the mine to give the mine’s volume. This approach was used at thirteeen sites. At the four sites that were surveyed by using a tape and clinomiter to construct cross sections, the area was determined by multiplying the area of each cross section by the lenghth it represents.  Survey Results  Surveyed mean mine site depths ranged from 0.6 m to 9.4 m.  The mean depth for surveyed sluice sites was 3.8m, for groundsluice sites was 3.8 m and for hydraulic sites was 6.6 m.  Tables 3.3a-d list all surveyed sites, details of the survey methodology and summary results.  Appendix A displays contour maps of all surveyed sites along with the calculated depth of removed material.  In addition to providing information on a surface area-volume relation, this survey data may provide valuable base information for industrial and/or historical-archaeological investigation.  Table 3.4 shows average values by mine type.  Both average area and average volume variables are very positively skewed.  When log transformed they are approximately normally distributed (figure 3.1).  Several exemplary or otherwise illustrative mine sites are shown in figure 3.2.  The topographic surveys that were used to construct Figure 3.2 are shown in full in Appendix A.  The site two kilometers north of Leon Creek (2 k N. Leon) is an example of a compound site with multi-layered history showing both classic sluice and hydraulic morphologies and where construction of the pre-mining surface was a straightforward exercise.  The upper portion of the site has classic sluice site morphology, one to three meters of sediment are excavated from the surface of a terrace.  Scarps around the mine and some “buttes” of unmoved sediment provide clear indications of the depth of sediment removal in this portion of the mine.  An unorganized cobble lag covers the floor of this part of the mine.  Three large hydraulic pits cut up to 20 meters into the terrace.  This probably removed much of the part of the site that had been sluiced.  Reconstruction of the original surface over the hydraulic part of the mine was facilitated by interpolation between clearly preserved facets of the pre-mining terrace scarp.  The line of intersection between the original terrace surface and the original terrace scarp surface was determined by connecting the high points defining the terrace edge between the hydraulic pits.  To the south of the hydraulic pits, two gullies run down to the river below the mine.  These gullies may have been incidental, resulting from erosion associated with outwash below the sluice site on the terrace above, or may have been intentional prospecting efforts associated with the hydraulic phase of mining at the site.  Such gullies are not uncommon below mine sites.  The Big Bar hydraulic site is cut into two different terrace levels.  The lower terrace, where approximately 5 m of material were removed, was worked first. The original surface over the lower excavation was defined by extrapolating the un-mined lower terrace to the west over the mined 55 Name Easting Mine Type Survey Type unnamed 534681 5862484 sluice 6366 25464 4 NA 535406 5861808 hydraulic 107964 794615.04 7.36 NA 11 Mile Flat 537029 5855632 sluice 190415 1207231.1 6.34 NA 536731 5852391 35794 137448.96 3.84 NA unnamed 535053 5839629 4392 28328.4 6.45 NA Bullion Pit 592209 5831550 hydraulic 388910 14634683.3 37.63 Rocky Point West 535687 5821914 sluice 26468 119106 4.5 NA Rocky Point East 536004 5821762 sluice 29006 130236.94 4.49 NA W. Onion Bar 550156 5705121 sluice partial survey 57954 259054.38 4.47 TIN, manually edited Big Bar Ferry 560419 5670498 complete survey 28583 168639.7 5.9 Big Bar Hydraulic 562184 5669938 hydraulic complete survey 140210 974459.5 6.95 TIN, manually edited 1/2 km S of Ward Creek 567214 5663442 sluice partial survey 59276 222285 3.75 TIN, manually edited Wheelbarrow Flat 567982 5662635 sluice complete survey 37933 70176.05 1.85 TIN, manually edited 1 k Above High Bar 570643 5661315 complete survey 17512 57089.12 3.26 TIN, manually edited W. High Bar GS 571183 5661257 complete survey 2344 4031.68 1.72 TIN W. High Bar Sluice 571252 5660962 sluice 25293 71579.19 2.83 TIN, manually edited High Bar NE 571954 5660686 98045 550032.45 5.61 NA Northing Area (m2) Volume (m3) Average Depth (m) Original surface interpolation method point depth estimates 7 mi S Quesnel point depth estimates point depth estimates Kersely Creek groundsluice point depth estimates groundsluice point depth measurements DEM from geobase 60% global polynomial interpolation point depth estimates point depth estimates sluice/groundsluice 75% global polynomial interpolation sluice/groundsluice groundsluice partial survey, extrapolated to whole site sluice/ground sluice complex point depth estimates Location Table 3.3 a): List of Mine Site Surveys 56 Name Easting Mine Type Survey Type 2 k N of Leon 575195 5652688 hydraulic/ sluice complete survey 56533 390643.03 6.91 Upper Leon 575272 5650910 complete survey 19257 115734.57 6.01 Lower Leon 576026 5650808 complete survey 56582 267067.04 4.72 TIN, manually edited 579791 5638627 sluice complete survey 1283 782.63 0.61 579827 5638445 sluice complete survey 1589 1001.07 0.63 580191 5638384 sluice 13819 73240.7 5.3 N/A 579901 5638327 complete survey 2199 4156.11 1.89 TIN 579847 5638253 sluice complete survey 5934 11630.64 1.96 Bridge River "Big Bend" 559486 5634758 hydraulic 199059 15307637.1 76.9 E Lee Hydraulic 579345 5629641 hydraulic complete survey 45119 242289.03 5.37 E lee a) 579181 5629603 ground sluice/chute complete survey 1950 4348.5 2.23 TIN E Lee b) 579046 5629538 Sluice complete survey 6366 9549 1.5 TIN E Lee c) 579143 5629514 sluice complete survey 21468 8123.71 2.81 TIN, manually edited 580265 5622481 hydraulic partial survey 15360 106695 4.97 TIN, manually edited Northing Area (m2) Volume (m3) Average Depth (m) Original surface interpolation method TIN, minimal manual editing ground sluice/hydraulic global polynomial interpolation hydraulic/ground sluice W Bank @ Skwish Creek a) 100% local polynomial interpolation W Bank @ Skwish Creek b) 100% local polynomial interpolation Skwish Creek point depth estimates W Bank @ Skwish Creek d) eroded slope below mine W Bank @ Skwish Creek c) global polynomial interpolation DEM from geobase global polynomial interpolation 50% local 2nd order polynomial interpolation Fountain Hydraulic Location Table 3.3 b): List of Mine Site Surveys 57 Name Easting Mine Type Survey Type 576554 5618104 hydraulic 28797 155503.8 5.4 TIN, manually edited 575910 5615988 hydraulic complete survey 12236 75129.04 6.14 TIN 575956 5615565 sluice 231379 712647.32 3.08 scarp heights Texas Creek Hydraulic 584620 5600981 hydraulic complete survey 22328 207203.84 9.28 TIN, manually edited Fosters Bar 589460 5596178 sluice 1928 8072.73 4.19 TIN, manually edited SIR5 589687 5595398 sluice complete survey 84556 339069.56 4.01 TIN, manually edited 592495 5589901 complete survey 2780 11092.2 3.99 592578 5589857 complete survey 4427 24791.2 5.6 592654 5589811 complete survey 1444 8664 6 592700 5589791 paced 96 576 6 NA 593097 5586126 complete survey 39429 121835.61 3.09 TIN, manually edited Cameron's Bar 594070 5583310 sluice 114183 773018.91 6.77 594155 5582433 complete survey 22369 70238.66 3.14 595561 5577519 82290 554039.11 6.73 NA Northing Area (m2) Volume (m3) Average Depth (m) Original surface interpolation method Lillooet Old Bridge Hydraulic 1/2 of site surveyed, average depth extrapolated to whole site Lillooet Mill Hydraulic Lillooet Mill total station point estimates survey - limited extent Laluwissin e groundsluice/sluice 50% local polynomial interpolation & manual TIN creation Laluwissin d sluice/groundsluice global polynomial interpolation Laluwissin c groundsluice/sluice global polynomial interpolation Laluwissin a sluice/groundsluice Siwhe Ground Sluice groundsluice manual survey: tape, level and staff assumption of a flat plane Izeman North sluice/groundsluice 50% local polynomial interpolation Splintlum Flat groundsluice/sluice point depth estimates Location Table 3.3 c): List of Mine Site Surveys 58 Name Easting Mine Type Survey Type 598285 5570296 hydraulic 22008 208855.92 9.49 TIN Rip Van Winkle Flat 599108 5569023 439211 1651433.36 3.76 NA 598882 5568720 hydraulic 13438 98366.16 7.32 NA 600999 5564186 sluice 23463 112622.4 4.8 TIN, manually edited North Bend 611463 5525745 sluice & hydraulic point estimates 207004 747284.44 3.61 NA Emory Flat 614656 5485726 Sluice 168811 1269458.72 7.52 NA American Creek 613541 5477207 sluice 14968 80976.88 5.41 NA 613884 5477066 sluice 8 cross sections 50209 180250.31 3.59 NA Northing Area (m2) Volume (m3) Average Depth (m) Original surface interpolation method Hydraulic N. of Rip Van Winkle Lytton Moran Survey sluice, limited hydraulic point depth estimates hydraulic west of Van Winkle Flat point depth estimates Lytton Railroad Bridge W. 95% survey, 5% visual estimation 3 cross sections, point estimates 4 cross sections Trafalgar Location Table 3.3 d) : List of Mine Site Surveys Sluice Hydraulic Depth (m) 3.77 4.71 6.57 64,840 18,587 44,270 289,736 97,682 295,292 Groundsluice Area (m2) Volume (m3) Table 3.4: Average geometry of mines by technology. 59 area.  Much of this surface is now buried in tailings from the excavation of the upper terrace.  Drains appear to have been intentionally cut through the edge of the upper terrace in order to provide a suitably low base level for excavation of the upper pit.  In some cases, large tailings fans are visible at the base of these drains.  Portions of one of these fans exist above the pre-mining surface of the original terrace.  Un- or partially-removed towers and buttes show a minimum elevation for the pre- mining surface at the river side of the upper pit and scarps bound three sides of the upper pit, which allowed for the pre mining surface to be easily reconstructed.  There are 1-3 m high piles of lag cobbles across the upper pit that are not resolved at the level of detail of the survey.  The sluice site ½ kilometer south of Ward Creek (1/2 k. S of Ward) is an example of a site Hydraulic and Groundsluice: log(volume) F re q u e n c y             Hydraulic and Groundsluice: log(area) F re q u e n c y               Sluice: log(area) F re q u e n c y               Sluice: log(volume) F re q u e n c y              Figure 3.1: Histograms of the logarithmically transformed variables area and volume for surveyed mine sites.  Sluice sites are displayed separately (right column), and hydraulic and groundsluice sites are stacked (left column).  Groundsluice sites are represented by the dark gray color. 60 2 k N. of Leon Big Bar Hydraulic 1/2 k S of Ward Siwhe Groundsluice ~100m ~200m ~100m ~100m Figure 3.2: Rendered views of some selected mine site surveys (left column of images)  and recon- structed pre-mining surfaces used to estimate the depth of excavations (right column). Scales are approximate because images are oblique perspective views. 61 where reconstruction of the original surface was very uncertain.  Two very different interpretations of the site’s history are possible.  Generally, the site is incised into the alluvial fan of a small creek and an associated river terrace.  It is possible that the whole site is incised into the fan and the original surface could be constructed by extrapolating the slight curvature of the fan over the whole site.  In this case 10 or more meters of material would have been removed from much of the site.  It is also possible that there was a lower terrace associated with the Fraser that was mined first and that secondary mining incised the upper terrace.  However, no remnant of the possible lower terrace is preserved at the site.   In this case, 2 or 3 meters of material were removed from the lower part of the site and up to 6 meters of material were removed from the upper part.  For the purpose of this work, the second, more conservative interpretation was adopted.  Assuming the macro picture of the second interpretation is correct, however, there were few details to guide construction of the pre- mining lower terrace surface.  Because of this, the high points of stacked lag deposits were assumed to be the elevation of that surface.  The estimate of material removed near to the river, therefore, is a minimum.  The intersection of the upper alluvial fan/terrace scarp and lower terrace surface was impossible to reconstruct in detail, and a best estimate of its location was determined.  A sharp break (unrealistic but probably reasonably accurate on average) then joined the lower terrace surface with the extrapolated upper alluvial fan/terrace surface.  Some other morphologic details that are interesting at this site include deep drains from the upper site leading to long sluice tailings piles and very thick lag deposits of large cobbles and small boulders in pits cut into the upper surface.  Much of the immediate post mining sluice drains on the lower part of the site have filled in with sand that was presumably deposited during floods of the Fraser.  Overall, therefore, the estimate for this site is a quite conservative estimate.  The Siwhe Groundsluice site is an exceptional example of the landscape effects of groundsluicing described by Kennedy (2009).  Steep scarps, level floors, and large heaps of cobbles behind walls of steeply stacked cobbles characterize the site.  A single large drain (11 m deep) services the whole site and irregular elongate pits are excavated in a fan above that drain.  These pits gradually get shallower with increasing distance from the drain.  Reconstruction of the pre-mining surface is a relatively straightforward task of connecting a wall on the river side of the excavation to the top of the scarps on the upslope side.  This provides a maximum estimate of possible material removed.  Because some natural drainage would follow the general shape of the site, it is possible that a depression existed prior to mining in the location of the site.  The depth of this depression was estimated based on high points in the mined area and the thicknesses of gravel lags.  The reconstructed surface is probably lower than was the pre-mining surface.  Similar interpretation and description could be narrated for all of the surveyed mines included in appendix A.  These few have been chosen to illustrate classic mine morphologies and the range of clarity left in the landscape regarding the form of pre-mining ground surfaces at mine locations. 62 Regression Relation Predicting Mine Volume From Mine Area  Because approximately ten percent of the identified mine sites were surveyed, regression relations predicting mine site volume from mine site area and type were developed as a means to estimate volume for any site where the area is known.  Using these calculations, it is then possible to produce a system-wide estimate of the total volume of mined material. The regression relations were developed using the statistical package R 2.9 (The R Foundation for Statistical Computing, 2009). In this instance it is appropriate to use regression, rather than functional analysis (Mark and Church 1977) because prediction is the object of the exercise. In so much as understanding an underlying physical relationship is concerned, regregression is also appropreate because the independent variable in the regression (mine area) is known much more precisely than the dependent variable (mine volume).  The variables mine area and mine volume appear, on the basis of limited data,  to be log- normally distributed (figure 3.1).  Therefore, in order to meet the assumptions of linear regression, both mine area and volume were log transformed before performing the regressions.  Three sites were removed from the regressions as they were known to be exceptional and exerted strong leverage.  These are the Big Bend Hydraulic site and the Bullion Pit mine, which are outside of the study region and were mined relatively recently using somewhat different hydraulic technology than was available for the surveyed sites; and “Laluwissin a”, which has an area three orders of magnitude smaller than the next smallest survey and which is smaller than nearly all of the mapped mine polygons.  To determine if there were differences in the trends between the mine types, a forward stepwise regression was performed predicting log(volume) by log(area) with mine types (hydraulic, groundsluice, and sluice) as class variables.  This showed that the trends for hydraulic and groundsluice sites are not distinguishable at α =0.10 (t= -1.260,  Pr>|t|) = 0.213).  Therefore, two separate regressions were developed: one with the groundsluice and hydraulic sites grouped and another for sluice sites.  Results of these regressions are plotted in figure 3.3.  For sluice sites the relation predicting volume (v) from area (a), after applying the transformation bias reduction method of Miller (1984)  is:                                                                        v=0.51 a1.21                                                                                (3.1) This regression is significant:  F= 321.6 on 1 and 26 degrees of freedom and P=3.688X10-16.  The assumptions of simple linear regression are met: There is no trend in the residuals and the residuals show equal variance (figure 3.4).  There is some possibility that the input values are spatially correlated, but there is no strong evidence that this is a problem.  The residual standard error for this regression is 0.2378 logarithmic units. Because the regression was performed in logarithmic space, the untransformed standard error is asymmetric about the predicted value and is plus 73% or minus 42%. 63  For groundsluice and hydraulic sites, the transformation bias corrected relation predicting volume from area is:                                                                        v=1.27 a1.166                                                                              (3.2) This regression is significant: F= 342.2 on 1 and 21 degrees of freedom and P=1.77X10-14.  The assumptions of simple linear regression are met: There is no trend in the residuals and the residuals show equal variance (figure 3.4). As with the sluice regression there is some possibility that the input values are spatially correlated, but there is no strong evidence that this is a problem.  The residual standard error for this regression is 0.182 logarithmic units.  As with the sluice regression the untransformed standard error is asymmetric. It is plus 52% or minus 34% of the predicted value.  The 90% confidence interval for the true mean of y given a particular x value (ŷ|xh) can be                        Log(area) L o g (v o lu m e ) Hydraulic Groundsluice Sluice Groundsluice and Hydraulic Regression Sluice Regression Symbols Represent Mine Type: Figure 3.3: Plot of regression relations predicting mine site volue from area for sluice (blue line) and groundsluice and hydraulic (green line) geomitries.  See text for details of the regression relations. 64                Fitted Log(area) Sluice Regression R e s id u a ls                  Fitted Log(area) Hydraulic and Groundsluice Regression R e s id u a ls Figure 3.4: Regression residuals for relations predicting mine volume from mine area for hydraulic and groundsluice (left panel) and sluice (right panel) and geometries.    	              
 
 
  	    Log (area m ) L o g (v o lu m e m )           3 2 Log (area m ) L o g (v o lu m e m ) 3 2 Groundsluice Hydraulic Figure 3.5: 95% confidence intervals (tight dashed lines) and 95% prediction intervals (loose dashed lines) for regression relations (solid lines) predicting mine  volume from mine area for hydraulic and groundsluice (left panel) and sluice (right panel) mine geometries. 65 calculated in order to give confidence bands for the regressions. For the sluice regression, this works out to                                               (ŷ|xh)±1.706 √(0.00202+0.00454(xh-4.33) 2)                                     (3.3) and for the groundsluice it is                                                (ŷ|xh)±1.721√(0.00143+0.00397(xh-4.18) 2).                                     (3.4) 90% prediction intervals for a new observation (ŷ(new)|xh) are                                              ŷ(new)|xh±1.706√(0.0586+0.00454(xh-4.33) 2)                                   (3.5) for the sluice regression, and                                              ŷ(new)|xh±1.721√(0.0345+0.00397(xh-4.18) 2)                                   (3.6) for the groundsluice regression.  The regressions, plotted with confidence intervals and prediction intervals are shown in figure 3.5.  Table 3.5 shows examples of the confidence and prediction intervals for both regressions across the range of mine site sizes.  Confidence intervals for the regressions range from  ± 10% of the predicted volume at the mean value to  ± 50% at extreme values.  Prediction intervals range from   ± 43% of the predicted volume to  ± 115% of the predicted volume. Sluice Predicted CI PI Predicted CI PI +% -% +% -% +% -% +% -% 1,000 1,658 26 21 81 45 3,261 42 34 115 73 10,000 26,767 12 11 75 43 47,811 17 16 96 62 15,136 77,526 16 15 94 61 21,380 67,025 11 10 75 43 100,000 432,116 16 14 76 43 701,035 26 22 91 58 1,000,000 6,975,896 32 24 85 46 10,278,980 47 36 96 58 Hydraulic and Groundsluice Mine Area (m2) Volume (m3) Volume (m3) Note: confidence intervals (CI) and prediction intervals (PI) are shown as plus or minus percent of the predicted volume. Italicized rows represent the predictions for the mean area for each regression. This is the location of the tightest confidence intervals. Because the regression was performed on log transformed data, the confidence intervals are asymmetrical about the predicted value in arithmetic space. Therefore, both the upper- and lower-bound intervals are explicitly presented. Table 3.5: Example 95% confidence and prediction intervals spanning the range of mine site areas for both the sluice and hydraulic and groundsluice regressions. 66 Conclusion  Results of the surveys and regression relation show that it is quite reasonable to estimate the volume of placer mines along the Fraser corridor based on their surface area.  The 95% prediction intervals have one-half to full order of magnitude spreads, meaning that prediction of the volume of any give mine is subject to errors on the order of  ± 100%.  95% confidence intervals for the regression, however, are much tighter (on the order of  ± 10-30%)  meaning that there can be a quite high level of confidence regarding summary predictions of total volume. Endnotes 1This particular sequence is somewhat common. It possibly represents distinct phases in the mining history of the region.  Sluicing activity was preeminent from 1858 through the 1870s or 80s; hydraulic activity was predominant as a large amount of capital flowed into the region in the ‘90s and beginning of the 20th century.  In the depression of the 1930s many individual miners without access to capital turned back to the region to try to make days’ wages and re-worked old sites with the small scale technologies of pan, rocker box, or sluices. 67 CHAPTER 4: ESTIMATING THE TOTAL CONTRIBUTION OF GRAVEL TO FRASER RIVER Grain Size Distributions  In order to determine the probable environmental impact of placer mining on the Fraser it is critical to know the grain size distribution of material that was dumped into Fraser River in addition to total volume of material that was dumped into the river.  Once released into the river, different grain sizes will have different mobility.  Sediment fine enough to be carried as suspended load would have mostly washed directly into the Strait of Georgia.  Sand will probably have fairly rapidly flushed through the canyon and over the steep gravel fan at the mouth of the canyon and have accumulated in the Fraser Delta.  Hales (2000) found high concentrations of mercury in Fraser delta sediment at depths dating from the 19th and early 20th centuries.  She speculated that this mercury was from gold mining.  Gravel and cobbles will have had a more complex interaction with the river.  These interactions will be explored in more depth in Chapter 5.  Any boulders that were dumped into the river have probably not moved very far from their point of entry.  Very little is quantitatively known about the grain size distribution of sediment that was placer mined.  Kennedy (2009) described cobble “tailings” – really lag deposits – left behind by the mining, which indicates the uncensored bulk mined material contained sediment at least as coarse as cobbles.  Each of the mining techniques employed along the banks and terraces above the Fraser had the potential to censor the material dumped into the river.  Placer gold is typically found in fairly small nuggets and flakes, so coarse material did not need to be extensively processed.  Hydraulic mining and groundsluicing use the erosive power of water to mobilize sediment and to wash it into sluice boxes.  Not all grain sizes are necessarily mobilized in the first place and some grain sizes may be intentionally diverted from the sluice boxes by hand. At sites that were primarily sluiced, material was loaded by hand into the sluice boxes and larger grain sizes were excluded, often leaving distinctive parallel stacked rows of cobbles.  Preliminary observation of lag deposits at mines revealed substantial quantities of boulders in addition to cobbles in the lag deposits.  Depending on the spatial  arrangement and ordering of operations at a mine some true tailings (material that was moved through a sluice box) may be left at a mine.  It is important to clarify some definitions. The material, as it existed before being affected by mining will be referred to as the uncensored material.  Material that never entered into the sluicebox and was left behind on the mine site will be referred to as “lag.”  Material that was dumped out of a sluicebox or otherwise washed away from a mine is tailings.  Narrative descriptions of grain size follow the Wentworth (Wentworth 1922) grain size classification except where gravel is used as a general term to denote deposits of mixed grain size that include gravel or coarser material.  The term “gravel-sized material” refers specifically to the Wentworth grain size range of 4-256 mm.  Figures present grain size in psi unit which are equivalent to the base 2 logrithm of the median grain axis in mm. 68 Methodology  Samples were taken of the uncensored material and the lag material from several mines. The difference between these is equivalent to the grain size distribution of the material that left the mine.  Notes were taken about the stratigraphy of each surveyed mine in order to ascertain the relative proportions of different facies.  Facies of the material affected by mining were defined and the proportion of material removed from the mine of each facies was estimated.  Bulk samples were taken of the uncensored material and Wolman counts were taken of the lag deposits.  Because bulk sampling is very labor intensive, somewhat invasive, and requires that heavy equipment be brought to the mine site, a limited number of samples was taken.  A more extensive database of similar earth materials from the region was compiled in order to place the samples in context and to determine reasonable sample ranges.  From these samples and the database of similar materials, representative grain size distributions were determined for each facies.  These distributions were integrated by the proportion of each observed facies to create a synthetic grain size distribution representative of the whole population of mines.  The heterogeneity of possible sequences was not apparent until late in field work.  Therefore, observations were not as systematic as would have been ideal.  They were based on either a visual estimation of the proportions of the scarp occupied by different facies, or by measurements of the scarp using a Jacobs staff.  Four facies dominated areas that had been mined.  The surface of most mines was mantled by loess.  Loess is wind blown material that is fine sand or finer.  It is easily recognizable and often has a reddish or brownish tint.  In some cases in the extreme northern and southern parts of the study region in the Spruce Sub-Boreal and Coastal Western Hemlock BGC zones, soil consisting of fine sediment (primarily loess though there may be components of colluvium or sand) and substantial amounts of organic material was grouped into the loess facies. The texture of this material is probably quite similar to loess.  Fluvial sand represents moderately well-graded deposits ranging in size from silt to granules. Fluvial gravel is a mixed-size facies consisting of fine sediment through gravel sized material, cobbles, or in some cases, boulders.  Clasts are rounded and the deposits are grain supported.  Typically beds of fluvial gravel are thick (greater than 1 m).  In a few cases material of glaciofluvial origin is included in the fluvial gravel facies. These deposits were identifiable by very steeply sloping beds.  Their grain size distributions and sorting characteristics appeared to be similar to the fluvial gravel facies.  Debris flow / colluvial deposits consist of angular clasts ranging in maximum size from gravel to boulders that may be either supported in a fine-grained matrix of clay-sand sized material or clast supported with a very fine-grained matrix.  These facies are fairly broad groupings. The fluvial gravel and debris flow/colluvial facies have a high degree of potential variability in the largest grain size.  At the beginning of sampling, it was assumed that the general grading characteristics of deposits with different maximum sized grains would be similar.  Therefore, information regarding the largest grains found was recorded as a modifier for the facies.  This was done either by a Wolman count of the associated lag or by visual estimation of the approximate D 90  69 of the material.  A total of twelve bulk samples was taken.  Locations of bulk samples were primarily determined by access constraints.  The sampling rig weighs more than was comfortable for two people to carry, so sampling locations needed to be near places a vehicle could reach. Within locations that vehicles could reach, sampling was targeted at deposits that seemed visually representative of each facies.  For facies with substantial variability, samples were taken over the range of conditions observed for the facies.  During initial fieldwork, composite, depth-integrated samples (Wolcott and Church 1991) were taken from the terrace scarp (and set of facies) at four sites. All of these included loess and only one other facies.  Because samples were integrated as volume- by-depth rather than weight-by-depth, the loess horizon typically represented 1/5th or less of the total scarp height, and loess has a low bulk density as compared to gravel, these composite samples can probably be considered as slightly diluted samples of the dominant facies.  Sample size was determined according to the 1% criterion of Church et al. (1987) up to a maximum sample weight of 500 kg.  Material was templated and weighed down to a size of 64 mm and then sieved and weighed down to a sub 22 mm sample, of which a split was taken back to the lab for further analysis.  The database of grain size distributions was collected from various published and unpublished sampling efforts that have been carried out in the region.  Sampling to produce all of these bulk distributions was carried out according to the protocol of Church et al. (1987) to varying sample size criteria.  Five unpublished grain size distributions were obtained from bars of the Quesnel river (Michael Church, personal communication 2010).   The samples were taken along the river from just below the Bullion Pit mine (where the main tailings were essentially sampled) to Quesnel Forks.  All samples in this set were collected to the 1% criterion.  Forty-nine bulk grain size distributions come from the gravel portion of Fraser River below Hope and were collected according to the 1% criterion.  Results from these sampling efforts are included in Church and Ham (2004). A set of 10 bulk distributions come from the terrace sequences at Lillooet and are reported in Ryder and Church (1986).  These samples were collected to at least the 5% criterion.  Finally, nineteen grain size distributions of debris flow material from the Southern Coast Mountains were obtained from Jordan (1994).  These were the only locatable grain size distributions reported from debris flow or colluvial material.  Wolman stone counts were conducted to sample mine site lag deposits at all except two sites where bulk samples had been collected, and at most sites where surveys were done. Approximately 100 stones were sampled (in some cases smaller samples from different areas of a site were aggregated) and recorded in half-phi intervals.  See Table B1 in Appendix B for details of each sample.  The distributions from Wolman counts give information on the maximum grain sizes present in the original unaltered deposit.  If information is also available about the grain size distribution of the original deposit, Wolman counts allow estimation of the sorting processes that occurred at the mine site. 70 Results Mine Site Stratigraphies and Facies Distributions  Figure 4.1 shows several examples of mine site stratigraphies.  Site stratigraphies are quite variable.  Approximately half (53%) of all surveyed mines are mantled by loess or soil.  In sites with multiple facies below the loess, sand, gravel, and colluvial or debris flow material may be interbedded.  We found that 15% of surveyed sites have sand, 34% of sites have colluvial or debris flow material, and 68% of sites have fluvial gravel.  Figure 4.2 shows box plots of the depth of material in each facies where that facies is present (values of zero were excluded).  The distribution of loess is highly asymmetrical.  In most cases, there was 0.5 to 1 m of loess.  At some sites, however, there were are deep (up to  3.5 m) deposits of loess.  In the few cases where sand was present it is typically 1 to 3 m deep.  The depth of colluvial deposits in mines is quite variable with most deposits being 1 to 6 m.  The depth of fluvial gravel typically ranges from 3 to 6 m.  It is possible to construct weighted averages of the proportion of material belonging to each facies at the mines that were surveyed by multiplying the total volume of the mine by the proportion of that mine that belonged to each facies (as determined by the average depth of that facies at the Surface -1m -2m -3m -4m -5m   
  	    -6m Debris Flow Sand Loess Boulder Gravel Cobble Gravel Legend Figure 4.1: Example mine stratigraphies. Note the loess cap and variable composition of sediment that was removed by mining. 71 mine) and summing the results by facies.  The resulting estimate is that 16% of the excavated material at surveyed sites was loess, 7% sand, 9% colluvium or debris flow, and 68% fluvial gravel. Bulk Sampling  Appendix B is a table showing comprehensive results of the both bulk and Wolman grain size distribution sampling.  Comparing the relatively few samples from deposits that were directly mined with other samples from the same facies in the region allows for evaluation of the representativeness of the set of samples taken in this study.  This will be done for the fluvial gravel and debris flow/ colluvial facies.  Because the fluvial sand and loess facies are both fairly minor components of the total volume of sediment and because they are almost completely fine enough to behave as wash load in the river, single samples of each suffice to generally characterize the facies for the purpose of assembling a synthetic Grain size distribution. Figure 4.3 shows cumulative distribution plots of all bulk mine site samples.  The dominant facies removed by placer mining was fluvial gravel.  Figure 4.4 shows cumulative distribution plots of samples from the fluvial gravel facies superimposed on plots of grin size distributions from the lower Fraser, Fraser at Lillooet, and Quesnel rivers.  With the exceptions of the Lillooet Mill and SIR5 Back Channel gravel samples, bulk distributions from mine site locations lie within the range of other samples of the facies for the region.  Both the Lillooet Mill and SIR5 Back Channel gravels were observed to be anomalous during field work.  Though they should perhaps be represented as a separate facies, similar material was not observed to be present in important quantities at other mine sites.  With the exception of the Lillooet Mill sample, mine site gravels are not bimodal like the lower Fraser gravels, but have similar minimum sizes and greater maximum sizes. Mine site gravels match the grain size distribution of the Quesnel gravels quite      Loess Fluvial GravelDebris FlowSand D e p th o f F a c ie s (m ) Figure 4.2: Box and whisker plots of the depth of each facies. 72      	    
     $          $   #  !" !" "   Figure 4.3: Cumulative distribution plots plots of all bulk mine samples by facies.     
  	  $-&* .&4" ,.&  &* "- $-&* .&4" ))     '  % ** "( - 1 "(    + ( " &( (+" /  3! -0 (&  &2 %" &( (+ +" /  &(( +./"-. - &*" &/". +2"- -."- &((++"/ "-- ". 0".*"( * ./"-&.'  #/"- /%" *)" +#  )&*" .),(" &*!& /". /%/ /%" .),(" 2.   +),+.&/" !",/% &*/"$-/"! .),(" 2&/% +".. &(0/&+* +# /%"  &".    	 					  Figure 4.4: Cumulative distribution plots of the fluvial gravel facies superimposed on plots of grain size distributions from the lower Fraser, the Fraser at Lillooet, and Quesnel rivers. 73 closely.  They are generally finer than the gravels from the Lillooet terraces, which were recognized as notably coarse by Ryder and Church (1986).  Figure 4.5 shows the grain size distribution of the debris flow and colluvial facies compared with the distributions from debris flows in the Coast Mountains (Jordan 1994).  The debris flow and colluvial facies is less well sorted than the fluvial gravel facies.  Samples from mines lie on the fine side of the range of Jordan’s samples. Wolman Sampling  Lag deposits are very strongly sorted (figure 4.6). Cumulative distributions are typically truncated somewhere between 128 and 360 mm from where they drop very quickly to a tail of smaller (16-45 mm) material.  Figure 4.7 is a histogram of the maximum grain sizes observed for all sites where Wolman counts were done. The modal maximum grain size is between 512 and 1024 mm. Maximum sizes of stones in the Wolman counts range from 91 to 4096 mm.  Figure 4.8 shows grain size distributions based on Wolman counts compared with grain size distributions of the un-mined source material.  Lag deposits are much coarser than their corresponding bulk deposits.  Comparison of the largest stone counted in Wolman counts of lag deposits with the largest from corresponding bulk samples (figure 4.9) shows that: 1) there is a weak "-&. (+2+((01&( +-!*   (01&( -1"( * ./"-&.'  #/"- /%" *)" +#  )&*" .),(" &*!& /". /%/ /%" .),(" 2.   +),+.&/" !",/% &*/"$-/"! .),(" 2&/% +".. &(0/&+* +# /%"  &".     		  	     '2 &.%  + #  -!  - 3 ! " +*  '         $-&* .&4" ,.&  &* "- $-&* .&4" )) Figure 4.5: Cumulative distribution of bulk samples from mines of the debris flow/ colluvial gravel facies superimposed on debris flow samples of Jordan (1994) and bulk fluvial gravel samples from mines. 74     	     ' $ ( . &(    $ '    ' $ ( . ## Figure 4.6: Cumulative distributions showing results of Wolman counts on mine lag deposits. 64-91 91-128 128-180 180-256 256-512 512-1024 1024-2048 2048-4096 0 1 2 3 4 5 6 7 max lag grainsize (mm) n u m b e r o f s ite s n=24 Fluvial Gravel Debris Flow and Colluvial Figure 4.7: Histogram of the maximum grain sizes observed for all sites where Wolman counts were done. 75 Figure 4.8: Comparison of cumulative distributions of bulk samples from unsorted material and Wolman counts from corresponding lag deposits.             maximum grain size observed in bulk sample m a x im u m g ra in s iz e o b s e rv e d in la g d e p o s it 1:1 lin e Figure 4.9: Comparison of the largest stone observed in Wolman counts of lag deposits with the larg- est stone in corrosponding bulk samples.     
  	  !*#' +#1 )+#  #' * !*#' +#1 &&     $  " '' % * . %    ( %  #% %( ,  0 *- %#  #/ " (+,*+ * (%# %#'+ * -%$ +&)%+ +" %#'+ *(%%&' (-',+ #* +&)%+ * ," +& (%(*    	 					  76 relation between the two (R2=0.41, p=0.124) and 2) the largest stone in the Wolman count may range from being equivalent to the largest stone in the bulk sample to a full order of magnitude larger and is on average approximately half an order of magnitude larger (1.7 psi units)1.  Analysis Estimating grain size Distributions for Each Facies  Only one sample was taken from each of the loess and sand facies, so the estimated grain size distribution for each of those facies is equal to that of the single sample.  Determining a representative grain size distribution for the Fluvial Gravel and Debris Flow and Colluvial facies is more difficult.  This was done by integrating knowledge gained from bulk distributions and Wolman counts of the lag material.  The observed correlation between the maximum size of lag stone and the maximum size of stones in the bulk distribution allows the maximum grain size of the best estimate Figure 4.10: Cumulative distribution plots of representative grain size distributions for each facies. The upper end of the representative distributions are anchored by data from Wollman counts (see text for further description of method). (!) #&,&##*+!# &(% 	 #*+!# (+# Measured Values Estimated Values (!) #&,&##*+!# #*+!# (+#     
  	  (!% )!- ' !  !% ( (!% )!- $$    	 					  77 for each facies to be anchored.  The average maximum grain size of lag deposits for the fluvial gravel facies is 786 mm and for the debris flow and colluvial facies is 989mm.  Because of the observed bias of 1.7 phi units for the coarsest material as estimated from a Wolman count of the lag material, the maximum size that would be predicted from bulk sampling of each facies can be estimated. This comes out to 242 mm for the fluvial gravel facies and 304 mm for the debris flow and colluvial facies.  The rest of the distributions were estimated by visually fitting a reasonable line that approximated the median value for the observed curves.  Figure 4.10 shows cumulative distribution plots of these estimations.  The Fluvial Gravel facies is composed of 18% large cobbles, 16% small cobbles, 33% gravel-sized material, 29% granules and sand, and 4% silt and smaller sized sediment. The Debris Flow and Colluvial facies is composed of 2% boulders, 11% large cobbles, 11% small cobbles, 37% gravel-sized material, 27% granules and sand, and 12% silt and smaller sized sediment. A sensitivity analysis shows that these values are not inordinately sensitive to the particulars of the manual estimation (Table 4.1).  The percent difference between the estimated value and the largest and smallest observed value for the facies range from 18 to 44%; except for the extreme ends of the spectrum (boulders and silt), where very small proportions result in very high percentage variation. Absolute differences range from 2 to 12 % and are typically small for gravel sized material. Constructing a Synthetic Grain Size Distribution to Represent All Material Dumped Into the River by Placer Mining  Using the data presented above it is possible to construct a synthetic grain size distribution to represent the sediment that was dumped into the river.  This distribution must account for three factors.  The first is the proportion of the total mined sediment that belonged to each facies and the second is the grain size distribution of each facies.  These first two points have already been addressed and can be combined to estimate the overall bulk distribution of all sediment that was affected by placer mining along the Fraser.  Because coarse lag deposits were left behind by the mining, we know that the sediment that was removed from the terraces and dumped into the river was modified from the original distribution.  This leads to the third factor that must be considered: sorting processes that occurred during the process of mining.  This factor is addressed in the Fluvial Gravel Debris Flow Sand Loess Estimated Coarsest Finest Estimated Coarsest Finest boulders 14 0 0 0 0 0 cobbles 37 12 22 ± 3 29 23 0 0 gravel 32 49 23 38 7 0 granules and sand 17 37 45 33 88 52 silt and smaller 4 ± 3 1 2 3 7 5 48 0 ± 7 2 ± 2 34 ± 12 33 ± 9 37 ± 7 29 ± 10 27 ± 12 12 ± 7 Error margins for the estimated values are the average absolute difference between the estimate and the coarsest and finest samples at that grain size. Single values are shown for sand and loess because only one sample was taken to represent each of those facies. Table 4.1:  Percent by weight of grains in each facies that belong to various grain size classes. 78 following paragraph.    Because paired samples of bulk distributions of the unaltered deposit and Wolman counts of the lag deposit are available, it is possible to see how sorting affects the bulk distribution. Unfortunately, the volume of the lag deposits is not known, so it is not possible to algebraically determine the grain size distribution of the tailings from the mine.  Figure 4.6 shows how mine lag deposits are strongly truncated.  There seems to be little relation between the bulk grain size distribution and the point at which the lag deposit is truncated (see figure 4.8).  The lag deposits are consistently truncated below the size of large cobbles (128-181mm).  It seems reasonable to simply cut off the upper end of the grain size distribution of the unsorted material at 128 mm to produce an estimate of the grain size distribution of the tailings.  The resulting estimate of the grain size distribution for tailings from the Fluvial Gravel faces is 20±12% small cobbles, 40±13% gravel-sized material, 35±6.7% sand and granules, and 5±2.5% silt and smaller material; and from the Debris Flow and Colluvial facies is 13±6.61% small cobbles, 43±8.7% gravel sized material, 31±6.2% sand and granules, and 14±6.4% silt and smaller material.   Error margines are based on a conservative sensitivity test identical to that done for the untruncated bulk samples where the maximum and minimum values are based on the mean of the difference between the value for the estimated grain size distribution and the grain size distribution of the coarsest and finest observed samples.  The composite grain size distribution for all tailings, obtained by multiplying the proportion of material in  each grain size for each facies by the proportion of material removed that belonged to that facies (as estimated by the proportion of surveyed sediment that belonged to each faces) and summing these values for each grain size, is shown in  5.10.  Based on the synthetic grain size distribution the average distribution of tailings for all mines along the river is estimated to be 14±7% small cobbles, 32±9% gravel, 41±4% granules and sand, and 13±4% silt and clay.  Error estimates from the preceding paragraph are carried through into these estimates for the synthetic grainsize distribution for mine tailings.  These error estimates, however, do not account for undetermined errors in the estimates of which facies were affected by mining. The Total Contribution of Gravel to Fraser River Mine Volume Calculation Methodology  The two regression relations, in combination with the 456 mapped polygons of mine sites described in Chapter 3, allow estimation of the total volume of sediment excavated by mining. Areas of the mines mapped in Chapter 3 were calculated by use of the “Calculate Geometry” Tool in ArcMap.  The areas determined by this operation were then applied to equations 4.1 and 4.2, as appropriate for sluice, and ground sluice and hydraulic sites, respectively.     At forty one sites where surveys have been done the volume of the mine as determined by the survey was used rather than the regression relation.  Also, during the mapping process, some mines were observed to be unusually shallow for their extent and notes were taken regarding the estimated average depth (typically 0.5- 2 m)  of the mine.  This was done in order to produce a conservative estimate of the total volume 79 of gravel produced by the mining while including historically interesting but less geomorphically significant mines in the mapping effort.  A few of these sites are quite large and may have the potential to introduce substantial error into the estimated volume of gravel affected by the mining were they to be used in the regression relation.  Because of this, at fifteen sites the estimated depth of the mine was simply multiplied by the area of the mine to produce an estimate of the total volume of gravel affected at the site. Results  Physical evidence confirms the conclusion that was reached at the end of Chapter 2:  a very large volume of gravel was affected by placer mining.  The best estimate for the total volume of material moved in association with mapped (according to Chapter 3) mines along the main stem of Fraser River between Hope and the Cottonwood Canyon is 58,000,000 m3 .  The 90% confidence interval for this prediction based on all of the mapped mines is from 55,000,000 to 62,000,000 m3. The lower bound for the 90% confidence interval with mapped mines of the most uncertain category (class 3) excluded is 47,000,000; this value can be considered a nearly absolute lower bound for the total possible contribution of gravel from placer mining along the river (subject to the possible determination that some high confidence sites were actually not mines and possible error in the regression relation), while the upper bound is fuzzy with the probability that unmapped mines exist. Full results of the application of the regression relation, with the location of mine sites,  estimated mine volume, and mapping confidence, are presented in Appendix C.  From the best estimate of the total volume, the the contribution from sites with sluice geometry is 35,000,000 , ground sluice geometry is 8,000,000 and hydraulic geometry is 16,000,000.   Figure 5.11 shows the longitudinal distribution of these inputs along the river and figure 5.12 is a proportional symbol map of individual mines along the river that shows the same pattern.  The pattern of gravel inputs, unsurprisingly, generally follows the clusters of mining activity observed in Chapter 3.  Large volumes of gravel were processed by mining between Hope and Yale, at Boston Bar, between Lytton and twenty-five miles upstream of Lillooet, in the High Bar-Big Bar area, between Canoe Creek and the mouth of the Chilcotin, and from fifty miles downstream of Quesnel to the Cottonwood Canyon.  Along Fraser River, mapped mines affected an estimated 14,000,000 m3  of sediment in the Cariboo District, with 8,700,000 m3  in the Quesnellemouth Division and 5,600,000 m3  in the Quesnelleforks Division; 22,000,000 m3  in the Lillooet District, with 6,400,000 m3 from above Big Bar, 15,600,000 m3 between just south of Lillooet and Big Bar, and 1,800,000 m3 between just south of Lillooet and the boundary of the Lillooet and Yale districts at Fosters Bar; and 24,000,000 m3 in the Yale District, with 17,000,000 m3  from above Hells Gate and 6,500,000 m3 below Hells Gate. 80 Comparison With Historical Information  Quantitative Historical information (see chapter 2) is available for a broader region than detailed physical observations associated with this study.  Furthermore, the historical record provides information on the timing of gravel inputs (see table 2.4 and figure 2.3).  Comparison between the physically observed impact of mining with estimates of the impact based on gold production values facilitates evaluation of the usefulness of the historical proxy record.  Table 5.2 shows estimates of gravel volumes affected by mining from both the observed physical record and estimates from the gold production value proxy. In areas where the volume of gravel can be estimated by both methods, the two are generally comparable.  In all cases  the observed volume of sediment affected by mining is greater, but by no more than one order of magnitude, than the estimate based on the gold production proxy.   This comparison indicates that estimates of the total sediment affected by mining made by the gold production proxy method as outlined in Chapter 2 are reasonable, but consistently negatively biased.   This is not unexpected because the time series sums are based on limited periods of time and necessarily do not include all activity that occurred on the river.  In case comparison of the physical evidence of placer mining with the proxy record based on Haggan’s (1923) 	  	    	 	    
   	   500,000 1,000,000 1,500,000 2,000,000 2,500,000 3,000,000 3,500,000 g ra v e l in p u t/ 1 0 k m o f ri v e r m 0 100 200 300 400 500 0 river distance upstream of Hope (km) 3 Figure 4.11: Longitudinal distribution of gravel inputs to the Fraser.  Values are summed for 10km long computation cells. 81 USA  Big Bar Quesnel Hope Lytton Lillooet YaleVancouver Williams Lake Boston Bar Aggasiz Mission Barkerville 
  
           Quesnell Forks Fraser River Lillooet R ive r Bridge R ive r C h ilcotin R iver Q u e sn el R iver Cottonw o o d R iv e r             	
    
   
  Figure 4.12: Proportional symbol map of estimated mine excavation volumes along Fraser River between Hope and Cottonwood Canyon. 82 gold production values, which would have accounted for nearly all of the placer mining activity, was possible. For this case, the proxy based estimation was within 15% of the physically observed value.  Using the gold production proxy to extend the estimates of gravel production beyond the Fraser seems to be reasonable, at least to give a basic picture of the magnitude of activity.  This extension shows an amount of sediment affected by placer mining along the Quesnel River  to be on the order of  fifty to one-hundred million m3. Less than 1 million m3 may have come directly from the Cottonwood and Swift Rivers, but very large amounts of sediment may have been affected by mining on smaller streams higher in the Cariboo that are sediment tributaries to the Fraser. Fifty-four million m3 are estimated to have been affected along Lightning Creek, a tributary of the Cottonwood River, while an additional  fifty-four million m3 are estimated to have been affected along the Willow River and tributaries which eventually discharge into the Fraser above Prince George.  The historical record also provides information on the timing of gravel inputs and different types of activity.  Figure 2.3 summarizes what is known about the timing of gravel inputs. Two peaks in placer mining activity are evident.  One peak around 1860 is associated with the initial rushes and largest number of individuals engaged in the mining, and the other, in the late nineties, is associated with capital investment and hydraulic mining.  Based on what is known of the historical record, Cottonwood River 700,000*** NA NA NA 54,000,000 NA 33,200,000** 89,400,000 NA Fraser in Cariboo 3,500,000* 9,500,000 14,000,000 Lillooet District 10,800,000 NA 22,000,000 Yale Division 6,800,000 NA 24,000,000 Estimated total sediment contribution based on gold production values from: Physically observed along the Fraser Time Series Sum (1858-1860 & 1874-1909) Haggen's (1923) Estimate Lightning Creek‡ Quesnel River and Tributaries * ** *** For the years 1872-1878 & 1882-1895 A tributary of Cottonwood R. Yale Division includes Thompson River For the very restricted time period 1882-1890 when records at this resolution are available Estimated by subtracting the value known to have come from the Fraser from the value estimated for the whole of the Quesnelleforks/Keithley Divisions and Quesnellemouth Division ‡ Lillooet Division includes Bridge River and Cayoosh Creek (Values are in cubic meters.) Table 4.2: Various estimates of sediment excavation associated with placer mining. 83 nearly all of the hydraulic mining was probably done in the eighties through the first decade of the 20th century, with some limited activity in the 1920s and 1930s. It is possible that some hydraulic mining, particularly in the region between Lillooet and the Cariboo, occurred during the period from 1862 to 1874 when little documentation of mining activity is available. Hydraulic technology was well established in California by the 1850s (Limbaugh, 1999) and so would have been in the toolkit of the first miners to enter into British Columbia in the ‘58 rush.   Even though greater volumes of sediment were estimated to have been affected by hydraulic mines in the preserved physical record than in the proxy based on gold production value, the volume of sediment attributed to hydraulic mining based on estimation from gold production values along the Fraser in both the Lillooet District and Yale Division accounts for a much greater proportion (78 and 74%, respectively) of the total sediment than that observed in the preserved physical record (25 and 17%, respectively), so it does not seem likely that a large amount of undocumented hydraulic mining occurred.  Fifty-nine percent of sediment generated by placer mining came from sites that were classified as having a “sluice” morphology.  This dominance raises an interesting question regarding the historical understanding of placer mining in the region.  Estimates of the impact of different kinds of mining indicate that hydraulic mining was by far the most important in terms of sediment moved, which is in stark contrast to the observed situation.     It is possible that some “sluice” sites 100 1,000 10,000 100,000 1,000,000 10,000,000 0 20 40 60 80 100 0 20 40 60 80 100 120 Mine Volume m % C u m u la ti v e V o lu m e o f S e d im e n t N u m b e r o f M in e s Line Represents Cumulative Volume of Sediment Bars Represent Number of Mines 3 	 	  
	 Figure 4.13: Magnitude-frequency distribution of mine site volumes. 84 were worked by low head hydraulic mechanisms and were able to profitably work pay values much lower than was estimated for sluice sites in Chapter 2.  Regardless, the dominance of sites with a sluice morphology along the Fraser indicates either that the presumed geomorphic dominance of hydraulic mining (e.g. Rohe 1986) is questionable or that the history of activity along the Fraser is exceptional.  Another way of viewing the relative geomorphic importance of organized and capitalized ventures over smaller-scale operations is to assume that larger excavations tended to be associated with larger and/or more organized operations and then to examine the magnitude-frequency distribution of placer sites.  Figure 5.13 shows the magnitude-frequency distribution of the volume of sediment affected by mining and the number of mines in half-order of magnitude volume classes. The greatest number of mapped mines are around 100,000 m3 but the greatest amount of sediment was excavated in mines of about one million m3.  More than 50% of sediment was generated by mines smaller than 312 million m3. Endnotes 1: The large difference between the largest stones encountered during bulk sampling and Wolman sampling of the lag deposit raises an interesting methodological issue relating to the grain size distributions of gravels and  bulk sampling in general.  The disconnect between the largest stones observed via Wolman counting of lag deposits and bulk sampling of unsorted deposits shows that, in the kinds of gravels that were sampled, there are quite a few stones that are much larger than would be observed in any reasonable bulk sampling program and that may not be observed even by large Wolman counts of the material.  Though they do not occupy the majority of the volume of sediment, such rare stones could have important implications for the development of armour layers, the overall erodibility of the gravel, and total flow resistance to water flowing over the deposit. 85 CHAPTER 5: THE FATE OF EXCAVATED GRAVEL  Once excavated during the mining process, sediment was either rejected immediately because of large size and left near its original position on the mine site, or it was passed through a sluice box.  At the downstream end of the sluice box a slurry of sediment and water was discharged and in order for the sluice to continue operation, there had to be sufficient dumping room for the quantity of sediment flushed through the box.  The most straightforward way to achieve this was to discharge the sediment directly into a body of water that could wash it away (Figure 5.1).  During field work, some tailings were still visible at many mine sites, but these tailings were typically piled on the surface of lower excavations in such a way that, in the course of survey, the volume occupied by the tailings was, effectively, subtracted from the overall volume of the mine.  In three cases, mapped mines were situated far enough away from the River to indicate that the volume of sediment excavated at the mine could not have entered the river.  The volume from these mines was not included in the estimates of the total impact of the mining presented in Chapter 4.    In one other case (the mine on the left bank of the Fraser approximately 11.5 km S of Quesnel), a large but unquantified volume of tailings is perched on a bench below the mine (Figure 5.2).  The total volume of this site was included in the estimation of the impact of placer mining.  Once in contact with the river, placer waste sediment will either remain in place, partially or fully damming the river, or be moved.  This chapter will consider the ways in which placer Figure 5.1: Tailings Dump of the Bullion Pit Mine into the Quesnel River. BC Archives Photo E-04753. Mcfetridge, F., 1935. Figure 5.2: Part of the large volume of tailings stored below the hydraulic mine located on the left bank of the Fraser 11.5 km S of Quesnel. 86 waste may have been conveyed through the system.  Key components of this assessment include description of the grain size distribution of sediment that the river is capable of moving, the capacity of the river to move the load that was introduced, and the rate at which the river would move the sediment through the system.  The discussion is opened with descriptions of physical observations that give some indication of the grain sizes that are mobile in the Fraser, and that show substantial aggradation has not occurred on the middle Fraser, which directly implies that any added sediment has been conveyed downstream.  It then considers the body of literature addressing movement of sediment slugs in gravel-bed rivers and attempts to classify the Fraser slug according to commonly used paradigms.   It concludes with attempts to semi-quantitatively assess the potential for sediment transport down the river by applying a sediment transport function to estimate the capacity of the river and a regression relation to predict the virtual velocity of downstream translation of the waste. Physical Evidence  Observations of the banks of the river below mines and of a recent large input of sediment to the river indicate that sediment dumped into the river by mining has been mobilized.  There is no sign of excavated sediment on the bank of the river below most mines, though at many sites, very small (10 to 200 m3 ) relict fans still protrude into the river.  A Wolman count (n = 197) of one such fan below a hydraulic pit on Van Winkle Flat revealed the grain size distribution shown in Figure 5.3. The vast majority of the surface sediment of the fan (96%) was larger than 91mm.  The very large size of the surface sediment of this tailings fan indicates that  selective transport has removed a large amount of  smaller material from the fan.   The fan below the Lillooet Hydraulic Mine at the “Old Bridge” provides a vivid illustration of the construction and destruction of a tailings fan (figure 5.4).  Another indication of the Fraser’s capacity to mobilize sediment from large inputs is the erosion and transport of large volumes of sediment below a landslide that entered the Fraser at Kersley in 2007 (PEP 2007) .  Though a substantial body of sediment still lies in contact with the river, large bars have formed below the slide since the event took place (Darwin Baerg1, personal Communication, 2009).  Both Wolman (n=117) and bulk samples were taken in the summer of 2010 from one of these bars about 1 km below the slide.  The largest stone encountered in the Wolman count (figure 5.3) of submerged surface material was 181mm, although a stone of 512 mm was noted on the surface of the bar.  The D50 of the stones sampled in the Wolman count was 45 mm.   The bulk sample was taken from a very loose, probably highly mobile, transverse ridge that protruded above the surface of the bar. The largest stone captured in the sample was between 45 and 64 mm .  The D50 of the sampled material is 11.5 mm.  Taken together, these observations of the grain sizes present at the newly formed bar below the Kersley slide indicate that cobbles have been mobile in the four years between the slide and sample date and that gravel sized sediment is highly mobile in the river.  A bed-load sediment transport function can be used to estimate the competence of the Fraser to move the volumes of sediment that were dumped into the river and to confirm that the 87 grain sizes of placer mine tailings are mobile in the Fraser.  Sediment transport was calculated by applying the a uncalibrated version (because no calibration data are available) of the Wilcock and Crowe (2003) surface-based sediment transport function which explicitly treats sand in the formula and includes a nonlinear effect of sand on the gravel transport rate.  In order to quantify mobile grain sizes, the function was calculated for three grain size distributions (generalized mine tailings and fluvial gravel from Chapter 5, and the grain size distribution of Orchard Bar from Ryder and Church 1986), at two sites (the Marguerite and Hope gauging stations) , for four flow conditions (smallest 1086420-2 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 grainsize (psi) % fi n e r E ro d e d ta il in g s fa n b e lo w V a n W in k le F la t N e w b a r b e lo w K e rs e ly S li d e In p u t G S D = O rc h a rd B a r In pu t g sd =F lu vi al G ra ve l F ac ie s In pu t G SD =M in e Ta ili ng s Smallest Flood at Marguerite Largest Flood at Hope Predicted GSD of Mobile Sediment Observed (Wolman) GSD Vertical Bars Represent the selective mobility threshold. Figure 5.3: Cumulative grain size distribution plots of Wolman samples from an eroded and cen- sored tailings fan below a mine (Van Winkle Flat) and from a newly formed bar below a recent landslide, compared with plots of the grain size distributions mobilized by flood flow conditions on the Fraser. The methods for calculating sediment mobility are explained in the text. The selective mobility threshold is here defined as the threshold between grainsizes where a greater proportion of the mobilized sediment belongs a size class than the available grainsize distribution, and grainsizes where a smaller proportion of the mobilized sediment belongs a size class than the available grainsize distribution. See the text for further description. These data suggest that the gravel-size sediment is readily mobile at most flow conditions in the Fraser and that cobbles must occasionally be mobile in order to create the observed censoring of the fan below Van Winkle Flat.  The sediment transport calculations suggest that mobile grain sizes in the Fraser are not very sensitive to flow conditions, but that the partial transport threshold is sensitive to the grain size distribution. 88 Figure 5.4: Paired historical (1886, top) and modern (2009,  bottom) photos looking downstream from the LB of the Fraser above Lillooet. The stage of the river in the historical photo appears to be about 2 m higher than it was at the time the modern photo was taken.  In the historical photo, sluice box lines are visible on top of a growing tailings fan on the RB of the river just downstream of the Bridge.  The  highly eroded relict fan is visible in the modern photo.  Also of interest is stability in the location of French Bar, the large medial bar just downstream of the mine. The historical image is a detail of a photo housed in the collection of the Lillooet Museum that was made by A.W.A Phair. French Bar French Bar Active Tailings Dump Relict Tailings Fan 89 observed flood, mean annual flood, 10 year flood, and largest observed flood (table 5.1).  Further details of input variable specification can be found in the section “Sediment Transport Modeling to Estimate the Fraser and Quesnel Rivers’ Capacities” on page 95.  The grain size distributions of mobilized sediment are shown in Figure 5.3.  For the fluvial gravel and synthetic tailings input grain size distributions, the largest grain size that was mobile was the maximum grain size in the assumed grain size distribution at all flows. With the input grain size distribution of Orchard Bar, material larger than 256 mm was only slightly mobile at all flows.  Because the function predicts sediment transport for many grain size fractions, it is possible to observe whether it predicts selective transport and, if it does, where the threshold for selective entrainment is.  This can be done by comparing the proportion of each size fraction in the mobilized grain size distribution to the input grain size distribution.  If the ratio for a given size class between proportion of input grain size distribution to proportion of mobilized grain size distribution is less than 1 for a given size fraction, that size fraction is disproportionately mobile; if it is greater than one, that grain size fraction is disproportionately immobile.  The function predicts that, for all flows, partial transport occurs. The threshold between sizes that are disproportionately mobile and disproportionately immobile varies based on the grainsize distribution.  It is 8 mm for the synthetic tailings input grain size distribution, 22 mm for the fluvial gravel input grain size distribution, and 64 mm for the Orchard Bar grain size distribution.  Both the observations from the Kersley Slide bar and the approximations in the Wilcock and Crowe sediment transport function suggest that grain sizes of 128 mm and smaller are readily transported during common flow conditions of the Fraser.  This fits well with the observed truncation of sediment smaller than 256 mm on the fan below a hydraulic pit at Van Winkle Flat.  Another important observation indicating downstream passage of sediment that was dumped into the river by placer mining is the apparent stability of the river channel.  If sediment has been added to a stable channel that has not aggraded, then, based on the principle of preservation of mass, that sediment (or an equivalent volume from the streambed) must, necessarily, have been transported downstream.  French Bar, downstream of the mine site shown in the paired photos in Figure 5.4 has (60 year record) smallest annual flood 3160 5.92 3.05 41.22 0.4 mean annual flood 4565 7.04 3.71 49.02 0.45 ten-year flood 5750 7.85 4.19 54.65 0.48 largest annual flood 6510 8.32 4.47 57.94 0.5 Hope (98 year record) smallest annual flood 5130 8.65 3.95 46.69 0.43 mean annual flood 8666 10.26 5.63 55.38 0.56 ten-year flood 10800 11.24 6.41 60.63 0.61 largest annual flood 15200 13.24 7.65 71.44 0.67 Flow (cms) Average Depth (m) Average Velocity (m s-1) Shear Stress (pascals) Froude Number Margeurite Table 5.1: Hydraulic conditions at various flood flows at the Hope and Marguerite Gauging Stations. 90 not changed perceptibly between the two photos.  Two historical accounts suggust that mining in the Fraser watershed had not disturbed the channel of the main stem of the Fraser in a way comparable to that in in California streams (Gilbert, 1917).  The mining commissioner at Lillooet, in 1891 stated: In view of probable extensive hydraulic mining on the various benches of the Fraser River in this district in the near future it is, perhaps, not out of place here to refer to a similar method of mining in other lands, notably on the Feather, Yuba, and Sacramento Rivers in the State of California, where legislative action was invoked to put a final stop to it on account of filling up the rivers named with silt, and the practical destruction and flooding of unnumbered acres of, probably, the best alluvial lends in that state.  From personal observation I am clear in saying the same conditions do not exist here, and capitalists investing largely in substantial, permanent works for the conveyance of water for hydraulic mining purposes, in this district at least, need not take into consideration the probability in the near or remote future of being met with legislative enactments or judicial injunctions compelling them to stop work. (ARMM 1891 p. 573) An account from 1923 maintains the same theme indicating that major channel changes had not taken place over another thirty years: “In Cariboo [hydraulic mining] is not hampered by anti-debris laws, damage suits for injury to farm lands, or interference with navigation as in California” (Haggen, 1923).  Modern observation shows that several sites where dredging occurred still have distinct scars that are visible at low water (including the mouth of the Anderson River, perhaps High Bar and Horsebeef Bar at Lillooet (Kennedy 2009).  Early placer-mining-era descriptions of the locations of bars (see chapter 2 and Map 2.1), for the most part, match the locations of modern bars.  The only region where there are substantial, unstable bars indicating the presence of a large amount of sediment (and possible aggradation), is the part of the river from the upstream limit of this study (Cottonwood Canyon) to Soda Creek. Another indication of instability in the area is that there is no physical record of what seems to have been a very important mining area near the mouth of Narcosli Creek (13 Mile Flat), which indicates probable lateral instability of the river at that location.  Some of bars in this reach are clearly new and associated with the recent input of sediment from the Kersley Slide, but others have become forested islands and may be composed of placer mining sediment. In addition, there is substantial morphological evidence of aggradation in the Quesnel River, where a very large volume of sediment from placer mining was discharged. It is possible that some placer waste sediment from the Quesnel or other streams in the Cariboo is beginning to affect this reach. Furthermore, although no major aggradation is visible on the Fraser it is important to note that the volume of placer waste gravel in the river, if evenly distributed along 500 kilometers of the Fraser’s 250 m wide channel would produce only 17 cm of aggradation, which would not be detected by the coarse observations that have been presented.  However, considering irregularity in the gradient of the channel and constraining canyons, evenly distributed aggradation is highly unlikely. 91 Sediment Slug Behavior Introduction  Though it has been demonstrated that a large volume of sediment was excavated by placer mining, that most of that sediment was dumped directly into the Fraser, and that most of the grain sizes of material dumped into the river are mobile at commonly observed flows, the potential for downstream transport of placer-mining waste has not been evaluated.  Though it may be possible to numerically model transport of the estimated sediment inputs to the river, such work is beyond the scope of this study.  Fortunately, however, a broad body of literature exists addressing the phenomenon of sediment slugs.  A review of this literature will be presented to provide context for a qualitative assessment of the probable response to the sediment inputs to the Fraser associated with 19th century placer mining.  This qualitative assessment will include estimates of the competency of the river to convey the retrospectively predicted loading and estimates of probable virtual velocities of sediment in the Fraser.  This discussion of that literature will begin with a description of the different terms used to refer to the phenomenon.  It will proceed with a discussion of how sediment slugs have been classified and the ranges of behavior that have been observed.  It will then attempt to describe the degree to which a unified theory regarding the controls of sediment slug behavior has been developed by assessing the debate between the relative importance of translation and dispersion of sediment waves and a discussion of the role boundary conditions play in determining sediment slug behavior. Theoretical Background  At least five different labels – bed material waves, sediment waves, bed waves, sediment slugs, and sediment pulses – have been used to describe “systematic changes in channel-bed elevations that form an aggradation-degradation cycle” (James 2006).  Each term has specific connotaitons regarding the behavior of the phenomenon. Sediment slugs are “bodies of clastic material associated with disequilibrium in fluvial system over time periods above the event scale” (Nicholas et al. 1995 p. 502).  Sediment slug is preferable as a general term to describe the phenomenon because the degree to which transport occurs as a coherent wave has not been established. Sediment slugs were first described by Gilbert (1917), who chronicled the impact of a very large input of sediment from placer mining on the Sacramento River and it tributaries.  A very large slug of sediment resulting from placer mining resulted in a several meter aggradation- degradation sequence of the bed of these rivers in the Sacramento Valley.  Gilbert conceptualized the phenomenon as a downstream translating wave “analogous to the downstream movement of a great body of storm water” (p. 30).  This perspective has been the starting point for research into the phenomenon but many researchers have shown that the perspective is not universally applicable (eg Bartley and Rutherfurd 2005) or even necessarily applicable to the debris from hydraulic mining in California (James 1989, 1993, 2004, 2006) and that there is a large spectrum of behavior of sediment waves. 92 This spectrum of behavior has been classified in a variety of ways.   Sediment waves have been observed from the scale of individual bedforms to basin-scale valley floor adjustments. ‘Macroslugs’ are controlled by fluvial process-form interactions and generally have minor impacts on the fluvial system, ‘megaslugs’ are controlled by local sediment supply and valley-floor configuration and cause major valley change, and ‘superslugs’ are controlled by basin-scale sediment supply and result in major valley floor adjustments (Nicholas et al. 1995). The temporal persistence of a wave tends to increase with increasing spatial scale and range from annual variations to millennial persistence.  The sediment in a wave may come from within the system (endoslugs) if it is derived from sorting processes within the channel and riverbank erosion, or outside the system (exoslugs) if the  sediment comes from glacial outwash, mining debris, or slope input. Slope input is caused by destabilization related to forest clearing (Roberts and Church 1986, Madej and Ozaki 1996), forest fires (Germanoski and Harvey 1993, Hoffman and Gabet 2006), agricultural clearing (Meade 1982, Trimble 1983), and natural landslide events (Benda and Dunne 1997). Sediment associated with a wave may be stored only within the channel (eg Meade 1985) or on the floodplain (James 2006, Marron 1992, May 1982).  Larger slugs tend to extend their influence outside of the channel system and include substantial storage of material on the floodplain (Nicholas et al. 1995).  The grain-size of the material in a wave influences the behavior of the wave and is another basis for classification.  Cui et al. (2003a)  found that when the caliber of grains in the slug was similar to, or coarser than the bed material – the bed material serving as an indicator of the competence of the stream – it became armored and evolved primarily through dispersion.  When the caliber of the slug was finer than the bed material the slug evolved rapidly through both translation and dispersion.   Field studies have observed both translation, where the slug material is fine (Knighton 1989, Marron 1992, and Meade 1985) and dispersion of coarse slugs (Sutherland et al., 2002 and Hoffman and Gabet, 2007).  Translation and dispersion have been identified as primary propagation mechanisms for sediment waves.  Figure 5.5 depicts visual definitions of these propagation mechanisms. The situations governing the controls between these modes are discussed at length in Lisle et al. (2001) and Cui et al (2003a&b).   A third propagation process, that of a “migrating inflection point between aggrading and degrading zones,” is described by James (2006), who argued that this was the most Figure 5.5: Visual definitons of translation and dispersion. A) Dispersing Slug A) Translating Slug C) Translating and Dispersing Slug 93 appropriate paradigm for viewing the system that GK Gilbert (1917) described.  Lisle et al. (2001) found that, for waves in quasi-uniform, gravel-bed channels when exchanges of sediment between an channel and its floodplain are neglected, dispersion dominates the evolution of bed material waves in gravel bed rivers.  They use the following form of the Exner equation, written in terms of Froude number to explain this observation : (5.1) This equation is based on the Meyer-Peter-Muller (1948) bedload transport equation. Parameters that are informative in the present discussion are n, which is bed elevation, t and x, which are time and distance, u and h, are fluid velocity and depth, and F, which is the Froude number.  The terms are for wave dispersion and translation, respectively.  We observe that, as F approaches 1 (critical flow) the term for translation, goes to zero.  Lisle et al. (2001) proceeded to describe how in rivers with low Froude number, translation may be important but that “steep, self-formed, gravel-bed channels tend to have high Froude numbers during high discharge.” This conclusion is supported by the preponderance of observations of sediment slugs in natural gravel bed rivers (Sutherland et al. 2002, Hoffman and Gabet 2006) and flumes mimicking gravel bed rivers (Lisle et al. 1997, Cui et al. 2003 a&b). The translation-dispersion paradigm as been developed only for quasi-uniform, gravel-bed channels, while neglecting exchanges of sediment between a channel and its floodplain, and therefore may not be appropriate for many large-scale sediment slugs.   There are many situations where floodplain storage is important and many heterogeneous channels, therefore examples of sediment slugs that are not easily fit into the paradigm are abundant. Examples of waves with significant overbank storage can be found in James (2006), Marron (1992), and May (1982).  Examples of heterogeneous channels impacting the passage of a sediment wave are Wathen and Hoey (1998) and Lancaster et al. (2001).  Wathen and Hoey (1998) studied an example of a sediment wave that caused local instability leading to amplification of the sediment wave and proposed that local morphology and exchanges between the wave and channel have a critical impact the propagation of sediment waves.  The sediment connectivity paradigm is a helpful framework within which to consider any discussion of sediment waves in heterogeneous channels. Hooke (2003) proposes that there are many situations in which coarse sediment sources and sinks are local and that any given particle will not move through the system in the timescales of sediment waves.  She proposes that channel reaches that lack bars either 1) lack competence to move material through the reach (leading to an unconnected system), 2) flush the sediment due to high competence (leading to a connected system), and 94 or 3) lack sources of sediment (leading to a potentially connected system), and shows that each of these situations can be discerned by the morphology on either side of the reach that lacks bars. Applying this concept to translating sediment slugs would imply that they may move very rapidly through highly-competent reaches as undetectable low amplitude waves that become concentrated in sedimentation zones (Church 1983).  A general theory predicting sediment wave behavior may now be put forward.  The following situations promote translation:  Froude Number much less than 1, caliber of sediment in the slug is small relative to the caliber of sediment the river is competent to move, and situation of a wave at the prograding front of a sedimentation zone; while these factors promote dispersion: supply-limited bedload transport , presence of selective transport and abrasion, large wave material grain size, and high Froude number (Lisle 1997, Lisle et al 2001, Cui et al (2003 a&b).  The temporal and spatial scales of sediment waves are generally linked.  In addition, endogenous waves tend to be smaller than exogenous waves.  Finally, it is important to consider complexities of the channel network and the potential for exchanges of sediment within the system while considering sediment slugs. Implications for Fraser River  In order to use this information to better understand probable behavior of sediment that was dumped into the Fraser by placer mining it is useful to classify the slug and to determine the values of key parameters identified in the previous review such as the boundary conditions on the river, Froude numbers for various flow conditions and locations along the river, the grain size of both the bed of the river and sediment that was dumped into the river, whether there is selective transport of sediment in the Fraser and whether sediment transport is supply or transport limited. Classification of the Fraser Slug  The perceived scale of the slug on the Fraser (and tributaries) depends substantially on the scale and locality where it is viewed.  Inputs from individual mines are typically between 103 and 106 cubic meters (figure 5.13). In terms of scale then, individual inputs are mall functionally macroslugs.  To the extent that these individual inputs have coalesced, the wave is composed of up to 50 to 250 million cubic meters of sediment and is definitely on the order of several million cubic meters, which qualifies as a megaslug.   In order to be a superslug/Gilbertwave the wave must cause substantial valley floor adjustments and exist on the scale of the entire basin.  Where substantial mining occurred on smaller headwaters streams in the Cariboo, waves caused massive localized aggradation (Galois 1970) and would be classified as Gilbertwaves.  Because the wave is caused by mining sediment, it is an exoslug. There is no indication of over bank storage of sediment, except on small tributary streams in the Cariboo (Galois 1970), therefore the storage process is within channel.  Classification by grain size is based on the relation of the sediment slug caliber to the caliber of sediment on the streambed.  Though an estimate has been made of the grain size distribution of sediment that was dumped in the river by mining (see Chapter 5) very little is known about 95 grain sizes on the bed of the Fraser above Hope.  Ryder and Church (1986) report two grain size distributions for the bed of the Fraser near Lillooet (table 5.2) that are substantially coarser than the estimated “average” grain size distributions for sediment that was mined.  It is not possible to determine the wave propagation process for the placer waste slug through direct observation.  Using the list of criteria proposed on page 94, it is possible to use available data about the river’s hydraulics and geomorphology to make some inferences about probable behavior of the slug as it passes downstream from mine sites.  The Froude number along the river is typically well below 1.  At Marguerite, the Froude number during flood ranges from 0.4  to 0.48 and at Hope from 0.43 to 0.61 at the smallest and largest observed annual floods, respectively. Rapids are rare along the river, and typically occur in small parts of canyons, which occupy approximately 10% of the length of the part of the river of interest in this study. These generally low Froude numbers would indicate that translation is a possibility for transport of sediment in this very large gravel bed stream. The caliber of sediment in the wave is probably substantially finer than that of the river bed,  which would also suggest translation could occur.  However, the degree of mixing between placer waste and riverbed sediment is unclear.  Sediment transport is probably selective and supply-limited; these conditions would suggest dispersion.  Slugs generated by placer waste sediment have probably moved downstream on the whole (which necessitates substantial translation), but also, especially over the spatial and temporal scales of interest, have been substantially dispersed.  Regardless of whether downstream transport of material from individual mine inputs has been dominated by translation or dispersion, it may be contributing to aggradation. The key questions, then, are related to to the capacity of the river to move the load and the rate (virtual velocity) at which the river would move that sediment downstream.  The behavior of the slug in the aggrading reach is probably very different and has been studied in much greater detail (see Church and Ham 2004, Church et al. 2001). Quesnel --------(% fraction)------- 0.25 1 11 31 3 0.5 2 5 6 3 1 2 7 7 4 2 2 6 6 4 4 4 8 8 5 8 9 7 7 6 16 11 8 8 9 32 8 8 8 13 64 10 14 13 22 128 30 16 8 25 256 11 10 7 512 10 Finer than (mm) Orchard Bar Fluvial Gravel Mine Tailings Table 5.2: Input grain size distributions for sediment transport calculations. 96 Sediment Transport Modeling to Estimate the Fraser and Quesnel Rivers’ Capacities.  It is possible to use the sediment transport function to estimate the capacity of the river to move sediment.  Sufficient data are readily available to calculate sediment transport at three locations of interest: the Fraser River at Hope and Marguerite and the Quesnel River near Quesnel (Water Survey Canada (WSC) stations 08MF005, 08MC018, and 08KH006 respectively).  The average annual bedload transport at these stations can be calculated by using annualized partial duration series of flows (average number of days/year of a given flow) for the period of record at each station of interest (figure 5.6).  Average channel shear stress was calculated for the range of flows represented in the partial duration series.  This was done by using the hydraulic geometry of the channels to define average depth for each flow condition (Marguerite, from Hicken (1995), at Hope, based on WSC data provided by David McLean, personal communication 2010, and at Quesnel River at Quesnel based on WSC data provided by WSC staff person Lynne Campo), ; and defining the slope.  The channel slope over 4 km centered at the gauging station at Marguerite is 0.00071 (on the basis of high-precision GPS data from a downstream raft transect. Colin Rennie, personal communication 2010).  The channel slope from Hope downstream 13 km to Ruby Creek is 0.00055 (D. McLean, personal communication 2010).  The water surface slope for the Quesnel River at Quesnel decreases with increasing flow, according to the relation:                                                                     s = 0.13 Q^-0.63,                                                             (5.2) (WSC rating curve, Lynne Campo personal communication 2010) presumably because of backwater effects associated with the confluence with the Fraser.  Using average channel shear stress should have the effect of causing underestimation of sediment transport, because the sediment transport 50 100 150 200 250 300 3500 2000 4000 6000 8000 10000 12000 14000 16000 average no. days per year of equivalent or greater discharge d is c h a rg e (c m s ) Fraser at Hope Fraser at Margurite Quesnel Near Quesnel Figure 5.6: Partial duration series for flows on the Fraser and Quesnel rivers. 97 function is nonlinear with respect to shear stress. The additional flux in high shear stress parts of the channel will outweigh reduced shear stress in low shear stress parts of the channel (Ferguson 2003).  The sediment transport calculations assume channel widths of 175 m at Marguerite (based on cross sections presented in Hicken 1995), 150 m at Hope (based on cross sections done by the Ministry of Transportation 30 m upstream of the bridge between 1983 and 2001 (Dave McLean, personal communication, 2010), and 180 to 340m, depending on flow conditions, for the Quesnel near Quesnel (based on survey data presented in Smith 1974) (figure 5.7).  The sediment transport function is highly sensitive to the grain size distribution in the bed of the river.  Unfortunately, very little is known about the grain size distribution on the bed of the Fraser above Agassiz and for the Quesnel River below Likely.  Because of this, sediment transport was calculated using three different grain size distributions that are plausibly of interest for the sites on the Fraser: the  estimated synthetic grain size distribution for mine tailings, the grain size distribution for the “fluvial gravel” facies, and the grain size distribution observed by Ryder and Church (1986) on a bar just downstream of Lillooet, and for two grain size distributions for the site on the Quesnel, the grain size distribution for mine tailings and the average of a set of four samples from bars on the Quesnel River upstream of Quesnel Forks (data M. Church, personal communication, 2010) (table 5.2).  With these input variables, the Wilcock and Crowe (2003) transport function predicts that the Fraser River is capable of moving an amount of sediment comparable to the average annual input 0 50 100 150 200 250 300 350 400 0 2 4 0 2 4 6 8 10 12 2 4 6 8 10 12 14 0 Fraser at Hope Fraser at Marguerite Quesnel Near Quesnel horizontal position (m) h e ig h t a b o v e lo w e s t p o in t in c ro s s s e c ti o n (m ) Figure 5.7 Cross sections at locations of sediment transport calculations. See text for description of data sources. 98 of approximately 1 million m3 during the period of mining and that the capacity of the Quesnel severely limits conveyance of mine debris (table 5.3).  Results of the capacity calculations using different input grain size distributions span a wide range. The way in which mine tailings interacted with the existing bed of the river controls which capacity estimate should be used to approximate transport rate of the mine tailings.  The transport calculation using the grain size distribution of Orchard Bar is probably reasonably representative of the natural transport rate in the river.  Because the grain sizes present on the bed of the river are mobile at common flows, mine tailings probably mixed into this sediment.  The transport rate for the tailings, then, is probably best estimated by the capacity calculated using the Orchard Bar grain size distribution (200,000 to 400,000 m3 a-1).  These estimates are similar to the influx estimated in the gravel reach of the lower Fraser (Church et al. 2001). It is also possible, however, that mine tailings could have moved over top of the coarse (and in this scenario largely immobile) natural channel sediment.  In this case, the very high capacity estimated using the mine tailings grain size distribution may appropriately indicate that transport of tailings was almost entirely supply-limited and the only constraint on downstream movement of the slug would be the sediment’s virtual velocity.  The calculated transport on the Quesnel river is quite small.  This is consistent with the observed aggradation on that river.  No transport calculations were attempted for the Cottonwood River or Lightning Creek, but it is probably reasonable to assume that the capacity of these small streams is very much less than the loading and that their delivery of mine waste to the Fraser is small. Virtual Velocity  If transport in the river is supply-limited, then the key constraint limiting connectivity between mines and the lower river is the rate at which sediment would propagate . downstream through the system (the virtual velocity of the sediment).  Nicholas et al. (1995) report that typical sediment slug migration rates range from 0.1 to 0.5 km per year but that slugs generated by mine tailings may move up to 1-5 km per year, although the physical basis separating the behavior of “mine tailings” from other slugs is unclear.  Beechie (2001) compiled observations of annual travel distance (Lb) of bed load sediment at 16 locations, five of which were in gravel bed streams. The widest stream included in the study had a bankfull width of ~75m and the largest drainage area was ~550 km2.    He found annual travel distance to be strongly correlated with, and equivalent to approximately 20 bankfull channel widths (r2=0.86, p < 0.001).   Beechie provided a physical ----all sediment---- ----gravel only---- Quesnel Hope Quesnel Hope 29,000 8,885,000 14,324,000 2,890 2,052,000 3,593,000 Fluvial Gravel 3,087 3,368,000 6,061,000 551 1,218,943 2,330,000 Quesnel 80 - - 17 - - Orchard Bar - 297,000 696,000 - 197,044 396,000 Input Grain Size Distribution Margurite Margurite Mine Tailings Table 5.3: Calculated average annual bedload transport based on various input grain size distribu- tions at three locations.  Units are bulk cubic meters. 99 explination for this relation by combining the established relations between bar spacing and virtual velocity and channel width and bar spacing.  Assuming bar spacing of 5 to 7 channel widths (on the basis of Leopold et al., 1964; Keller and Melhorn, 1978; and Richards, 1982) he concluded that Lb  in typical alluvial channels is equivalent to three or four bar spacings.  Because there is some physical basis for the relation, it may be reasonable to extrapolate the regression to the Fraser, which is a much larger river than was included in the regression.  At the least, there is no reason to expect that the trends observed by Beechie would reverse with increasing stream size, so 1 km, the upper end of observed annual travel distances, can be considered the lowest reasonable Lb for the Fraser.  Bankfull widths in alluvial reaches of the river are typically between ~150 and 300m.   If the regression relation is extrapolated for the bankfull widths on the Fraser, the predicted annual travel distance through alluvial reaches is between 3.1 and 6.3 km per year.  Canyons along the middle Fraser can be considerably narrower.  However, because of high slopes, deep water, and a lack of storage zones, it is probable that sediment is conveyed very rapidly through canyons. About 10% of the Fraser, in the section of interest, flows through canyons (table 5.4).  These estimates of the virtual velocity of sediment in the Fraser can be used to construct a very simple model of the movement sediment introduced by placer mining through the system that allows for a retrospective forecast of sediment delivery volumes to the aggrading reach of the river. Figure 5.8 shows the results from applying this model.  With an annual travel distance (Lb) of 6.3 km a-1, annual delivery to Hope ranges from 300,000 to 1,000,000 m3 and all sediment that was dumped into the main stem of the Fraser would have been delivered before the turn of the millennium.  With a Lb of 3.1 km a-1 annual delivery values range from 200,000 to 700,000 m3 and sediment from the main stem of the Fraser is delivered to Hope through the 21st century. A lb of 1km a-1, the Canyon Length (km) 1.3 2.8 Hells Gate 2.7 4.1 Great Falls (Lillooet) 1.0 4.5 0.8 Pavilion (White Canyon) 6.3 High Bar 1.0 French Bar 1.2 Churn Creek 13.5 Iron Canyon 1.4 English Bluff 2.2 11.9 Soda Creek Canyon 1.6 btw. Yale and Spuzzum Black Canyon blw Hells Gate Above Kanaka Fountain Just above Fountain Chimney Creek to Wms. Lake Creek Table 5.4 Lengths of important canyons along the Fraser below Quesnel. 100 lowest reasonable estimate for the Fraser, results in annual delivery of  50,000 to 300,000 m3  and persistence of sediment from the main stem beyond the year 2350. These delivery values compare very reasonably with the observed rate of aggradation of 171,000 to 229,000 m3 a-1 for the period from 1952 to 1999 on the lower Fraser (Church et al. 2001). Field observations associated with this study have been limited to the main stem of Fraser River.  Historical observation suggests, however, that 2-3 times the volume of sediment that was excavated by mines directly along the main stem of the Fraser was excavated from mines on tributary streams  (See table 6.1).  It is much more difficult to estimate the capacity and virtual velocity of sediment in the headwater streams into which much of this sediment was dumped.  It may be reasonable to make some virtual velocity calculations for the Quesnel River (shown as the lightly shaded areas in figure 7.10).  This estimate assumes an unlimited capacity, however.   The peak delivery value of placer waste to the Quesnel of 4 million m3 a-1 is much higher than the capacity of the Quesnel and even substantially higher than the probable capacity of the Fraser at Marguerite. Additionally, recent air/satellite photos of Quesnel River show substantial bar complexes that are suggestive of significant aggradation, which would indicate persistence of much of the sediment that was dumped into this river.  The difference between loading and capacity of the Willow and Cottonwood is probably even more extreme.  Without much more careful work to quantify sediment transport rates along these streams, it does not seem reasonable to conclude that their sediment discharge into the Fraser has been elevated as a consequence of placer mining. 1900 2000 2100 2200 2300 2400 1 4 0 $ " "'  $ (  
  ( $ ) !    #" %# !  0.5 	  	 !)$  	 !)$   !)$     	 !)$  	 !)$   !)$    " &! $%$ '%" '$ " !  #%$( $&#" Figure 5.8: Estimated Gravel Delivery to Hope assuming unlimited capacity and various virtual velocities as specified in the figure.  The assumption of unlimited capacity is probably not warrented for the Quesnel (see text for more discussion). 101 Endnotes 1Mr. Baerg is owner of Fraser River Raft Expeditions.  He frequently leads expedition-length raft trips down the river. 102 CHAPTER 8: CONCLUSION  This thesis grew out of suspicion that the current rate of aggradation in the gravel-bed reach of the lower Fraser River may be connected to historical placer mining activity in the Fraser watershed.  Two key issues were explored in the course of the study: the extent and intensity of placer mining activity, and the potential for connectivity between the locations were placer waste was dumped into the river and the aggrading gravel-bed reach.  Relatively little was known regarding the extent of placer mining activity in the Fraser watershed outside of the part of the Fraser near Yale where prolific activity was known to have occurred during the ‘58 rush, the Cariboo, and the part mapped by Kennedy (2009).  Historical- documentary study and field mapping demonstrate extensive mining activity along the Fraser from the town of Hope upstream to the Cottonwood Canyon, and historical-documentary study indicates extensive activity along the Quesnel River and its tributaries.  Historical sources suggest that the mining occurred in two primary pulses: one during the period from 1858 to the mid 1860s, and the other from the mid 1880s to approximately 1905.  Physical evidence of 456 mine sites, with an average area of  26,000 m2 per site, was identified along the Fraser.  Detailed survey of selected mine sites to estimate their excavated volume allowed for development of  regression relations predicting mine volume from mine area with 95% confidence intervals for the regressions of 10-30%.  Application of the regression relation to the mapped mines results in a total estimated  bulk volume of excavated sediment of 45,900,000 m3 with a 90% confidence interval from 42,400,000 to 49,400,000 m3.  The measured volume of sediment excavated by mining compares reasonably with the volume of sediment estimated to have been affected by mining by using gold production values as a proxy for sediment volumes (see chapter 2).  This research demonstrates the power of combined historical-archival and field-based investigations of historical anthropogenic geomorphic impacts.  Historical data provided indispensable guidance to direct field investigations and provided insight into the timing and kinds of mining activity that occurred.  The correspondence of the proxy based and measured estimates of excavated gravel suggests that it may be possible to estimate the geographic impact of placer mining in regions where the history of activity is well documented (cf. Koshmann and Bergendahl 1968).  The majority of sediment excavated by placer mining was moved from mines in which less than 500,000 m3 were excavated.  The fact that relatively small mines and non-hydraulic mines dominate the magnitude-frequency distribution in the study corridor challenges the assumption that the primary geomorphological consequences of placer mining result from large, heavily capitalized, and predominantly hydraulic operations  (cf. Rohie 1986), and should encourage  further study of the geomorphic impact of placer mining in regions that were mined but not necessarily massively impacted by hydraulic activity.  Furthermore, there is physical evidence of extensive mining activity in regions where historical records would suggest activity was extremely limited. Such regions tend to be quite remote and, based on the available records of activity, worked primarily by Chinese and 103 First Nations miners.  The grain size distribution of sediment excavated by placer mining was estimated through bulk sampling of material equivalent to that excavated from mines and from Wolman sampling of lag material left behind at mine sites.  The estimated grain size distribution of mine tailing includes 14 ±7% small cobbles, 32 ±9% gravel, 41 ±4% granules and sand, and 13 ±4% silt and clay.   Because silt and clay (in addition to some sand) act as wash load they have been transported into the Strait of Georgia and most sand probably overpassed the aggrading gravel bed reach.  This leaves a total bulk volume of 21,100,000  m3 of gravel that might contribute to aggradation of the lower Fraser.  Very little sediment is visible in small relict tailings fans directly below mine sites, and there is no evidence of aggradation along the middle Fraser.  From a sediment budget perspective, then, if the sediment that was dumped into the river has not been stored, it must have been transported downstream.  Semi-quantitative modeling of the transport of this sediment in the Fraser indicates that the rivers capacity is comparable to or greater than the loading of placer waste and that annual transport distances (virtual velocity) of the sediment probably range from 3.1 to 6.3 km a-1 .  Estimated annual sediment delivery values to Hope, the beginning of the gravel bed reach on the lower Fraser, range from 100,000 to 700,000 m3 a-1 and compare favorably with the estimated annual aggradation of 171,000 to 229,000 m3 over the period from 1952 to 1999 (Church and Ham, 2001).  The timescales of sediment delivery suggest that sediment would have been delivered to Hope beginning almost immediately and would have continued at least through the end of the 20th century and may continue through the year 2100 or later.  It is highly probable that the rate of aggradation observed by Church and Ham (2001) has been substantially influenced by placer mining and is not representative of the long-term past rate or future rate of aggradation.  These observations underscore the argument that it is critical to historically contextualize geomorphic process studies and engineering monitoring programs (cf. Church 1980, Baker 1988 James 1999, James 2010).  Though it is clear that placer mining has caused substantial sediment loading to the Fraser, other sources of bedload have not been quantified and the relative importance of placer sediment is not clear.  The pre-Anthropocene gravel transport rate is not known.  Furthermore, other anthropogenic loading has almost assuredly occurred.  Important landslides are known to have occurred in association with construction of the railroads (Evans and Savigny 1994).  Data regarding the frequency-magnitude distribution of landslides onto Highway 1 and the CP rail line between Hope and Kamloops suggest that the volume of sediment recruited through landslides is modest – on the order of 2000 m3 a-1 (Hungr et al. 1999).  Sediment dumps associated with road cuts and tunnels are in contact with the river in several places.  Logging, ranching, farming, and urbanization have occurred in a relatively small portion of the watershed but, nonetheless, may have altered sediment loading to the river to some extent.  This effect, however, is probably almost exclusively limited to fine sediment except where logging may have triggered landslides or where run-away flood irrigation failure has washed away terrace edges (several such cuts were noted below agricultural fields during field work). 104  If placer mining waste were the only source of gravel contributing to aggradation in the Lower Fraser, it could sustain the observed rate of aggradation for approximately 100 years.  Another way of viewing the impact of placer mine waste is to consider the total amount of aggradation that may be produced by the accumulation of placer mine waste.  This can be roughly calculated by dividing the total active channel area in the aggrading reach by the volume of placer waste. This produces an estimate of 0.17 m of average aggradation in the river’s channel, although the maximum local aggradation may be much greater. Church et al. (2001) report measured average aggradation of 2.1 cm in the gravel bed reach over 47 years, but that actual aggradation is strongly localized with many reaches experiencing up to 1.5 m of fill.  Observation of the sediment slug associated with placer mining on the Fraser provides limited new understanding in the study of sediment slugs.  From the observed evidence, the Fraser slug has not exhibited any abnormal behavior.  This observation is important because the Fraser slug exists on a much larger scale than any previously documented slugs.  If there is indeed a connection between placer waste and the observed aggradation in the lower mainland, then the Fraser slug demonstrates coalescence of a dispersed and non detectable wave into a detectable wave of geomorphic significance where the slope of a stream is reduced.  This behavior was hypothesized by Lisle et al. (2001).  Description of the ways slugs may interact with changing boundary conditions (eg. Hooke 2003), and of the virtual velocities of sediment in gravel-bed rivers (Beechie 2001) proved to be the most important tools for semi-quantitative assessment of the downstream movement of placer waste in the Fraser.  There is great potential for application of a detailed sediment routing model along the river to study behavior of the wave because the sediment budget is well constrained at both the input sites (with the exception of the natural load of the Fraser and tributaries) and terminus of the gravel bed reach.  This study suggests caution should be used when applying the measured historical gravel budget for the aggrading reach as forecast for the future and basis for management decisions.  There is a substantial probability of a regime change as the river finally digests sediment that was introduced by placer mining, making critical vigilant monitoring of the current movement of sediment into the lower Fraser and effects of ongoing sediment removal activities (Church, 2010). 105 WORKS CITED Alta California or Daily Alta California, (1858 and 1859), San Francisco, California, UBC Library, Koerner (microfilm), Vancouver B.C. Baker V.R., (1988) “Geological Fluvial Geomorphology” Geological Society of America Bulletin v. 100, pp. 1157-1167. Baker, R., (2001) “An outline of the mining history of New Zealand.” New Zealand Mining.  Vol 28. Bancroft, H.H., Nemos, W., Bates, A., (1887)  History of British Columbia: 1792-1887. The Works of Huber Howe Bancroft vol. XXXII.  The History Company, San Fransisco. Banyaurd, W. H.H.,(1891) Chief Engineer U.S. Army Annual Report for 1891. Bartley, R., Rutherfurd, I., (2005) “Re-evaluation of the wave model as a tool for quantifying the geomorphic recovery potential of streams disturbed by sediment slugs.” Geomorphology v. 64 pp. 221-242. Base Mapping and Geomatic Services Branch (1948), BC590/X170 1-11 [aerial photograph] 1:13,000 Victoria, B.C. Ministry of Agriculture and Lands Base Mapping and Geomatic Services Branch (1949), BC856 78-96 [aerial photograph] 1:40,000 Victoria, B.C. Ministry of Agriculture and Lands Base Mapping and Geomatic Services Branch (1951), S69-R4 1-24 [aerial photograph] 1:13,000 Victoria, B.C. Ministry of Agriculture and Lands Base Mapping and Geomatic Services Branch (1951), S84-R4 305-316 [aerial photograph] 1:13,000 Victoria, B.C. Ministry of Agriculture and Lands Base Mapping and Geomatic Services Branch (1951), S84-R1 720-730 [aerial photograph] 1:13,000 Victoria, B.C. Ministry of Agriculture and Lands Base Mapping and Geomatic Services Branch (1951), S84-R3 356-389, 521-582 [aerial photograph] 1: 13,000 Victoria, B.C. Ministry of Agriculture and Lands Base Mapping and Geomatic Services Branch (1952), S82-R1 665-680 [aerial photograph] 1:15,000 Victoria, B.C. Ministry of Agriculture and Lands Base Mapping and Geomatic Services Branch (1963), BC5070 163-185 [aerial photograph] 1:40,000 Victoria, B.C. Ministry of Agriculture and Lands Base Mapping and Geomatic Services Branch (1963), BC5071 11-15 [aerial photograph] 1:35,000 Victoria, B.C. Ministry of Agriculture and Lands Base Mapping and Geomatic Services Branch (2004), 15BCC04012 0082-0084, 15BCC04015 0157- 0158, 15BCC04015 0166-0167, 15BCC04016 0037-0038, 15BCC04017 0244-0245, 15BCC04018 0008-0009, 15BCC04018 0103-0105, 15BCC04018/0118-0119, 15BCC04021/0021-0023, 15BCC04021/0031-0032, 15BCC04028/0167-0168, 15BCC04036-0007, 15BCC04038/0054-0055, 15BCC04039/0168-0169, 15BCC04040/0014-0015, 15BCC04040/0076-0077, 15BCC04040/0092- 0094, 15BCC04040/0158-0159, 15BCC04041/0013-0014, 15BCC04043/0046-0047, 15BCC04043/ 0048-0049, 15BCC04043/0108-0110, 15BCC04043/012-0131, 15BCC04054/0091-0093, [aerial photographs] 1:20,000 Victoria, B.C. Ministry of Agriculture and Lands 106  Base Mapping and Geomatic Services Branch (2005), 30BCC05006/0036-0037, 30BCC05006/0098- 0099, 30BCC05006/0107-0108, 30BCC05006/0165-0167, 30BCC05008/0043-0044, 30BCC05008/ 0135-0136, 30BCC05031/0094-0095, 30BCC05031/0141-0142, 30BCC05031/0246-0247, 30BCC05032/0037-0038, 30BCC05032/0155 -0156,30BCC05032/0214-0215, 30BCC05033/0043- 0044, 30BCC05033/0115-0116, 30BCC05034/0144-0145, 30BCC05042/0027-0029, 30BCC05042/ 0130-0131, 30BCC05042/0183-0186, 30BCC05043/0055-0056, 30BCC05043/0091-0092, 30BCC05044/0007-0009, 30BCC05044/0106-0107, 30BCC05044/0184-0185, 30BCC05054/0075- 0076, 30BCC05054/0106-0107, 30BCC05056/0096-0098, 30BCC05061/0069-0070, 30BCC05069/ 0163-0164, 30BCC05070/0002-0003, 30BCC05074/0005-0006, 30BCC05074/0111-0112, 30BCC05074/0159-0160, 30BCC05074/0197-0198, 30BCC05075/0110-0111, 30BCC05075/0196- 0197, 30BCC05076/0012-0013, 30BCC05076/0134-0135, 30BCC05076/0149-0150, 30BCC05091/ 0069-0070, 30BCC05091/0075-0076, 30BCC05092/0171-0172, 30BCC05093/0119-0120, [aerial photographs] 1:30,000 Victoria, B.C. Ministry of Agriculture and Lands  Base Mapping and Geomatic Services Branch (2006), 30BCC06131/0048, 30BCC06132/0205-0206, 30BCC06132/0208-0209, 30BCC06132/0244-0246, 30BCC06133/0004-0007, 30BCC05067/027, 30BCC05066/267, 30BCC05029/114, 30BCC06004/151, 30BCC06004/009, 30BCC06003/068, 30BCC06030/023, 30BCC06062/108, 30BCC06120/108, 30BCC06119/069, 30BCC06119/060, 30BCC05067/028, 30BCC05067/026, 30BCC05066/268, 30BCC06029/113, 30BCC06029/115, 30BCC06004/150, 30BCC06004/010, 30BCC06003067, 30BCC06003/024, 30BCC06003/022, 30BCC06062/109, 30BCC061020/109, 30BCC06119/070, 30BCC06119/059, 30BCC06119/061, [aerial photographs] 1:30,000 Victoria, B.C. Ministry of Agriculture and Lands Beechie, T.J., (2001) “Empirical Predictors of Annual Bed Load Travel Distance, and Implications for Salmonid Habitat Restoration and Protection.” Earth Surface Processes and Landforms . 26. pp 1025-1034. Benda, L., Dunne T., (1997) “Stochastic forcing of sediment routing and storage in channel networks.” Water Resources Researchv. 33 no. 12 pp. 2865-2880. Bowman, A., [assisted by McEvoy, J.], (1887) Map of the Cariboo Mining District, British Columbia. Geological and Natural history Survey of Canada, Ottawa. Bradley, R.S., Jones, P.D., (1992) Climate Since A.D. 1500 Routledge. British Columbia Department of Lands and Forests Water Rights Branch, (1965) Sheet 8, Lytton Moran Survey. British Columbia Department of Mines (1946). Bulletin No. 21 Notes on Placer Mining in British Columbia. Charles F. Banyard, Victoria, BC. British Columbia Dept. of Mines and Petroluem Resources (1874-1910) “Annual Report of the Minister of Mine for the Year Ending...” The Department of Mines and Petroleum Resources, Victoria. British Columbia, Government Agent (Ass’t  Gold Commissioner ) Mining Records, (1860-1862), Alexandria District, [unpublished archival source] BCARS, GR-0216 Vol. 30-33 (ledgers), 107 Victoria, B.C. British Columbia, Government Agent (Ass’t  Gold Commissioner ) Mining Records, (1867-1899), Quesnellemouth Division, Cariboo District, [unpublished archival source] BCARS, GR-0216 vol. 76 (ledger), Victoria, B.C British Columbia, Government Agent (Ass’t  Gold Commissioner ) Mining Records (1859-1861), Lillooet, [unpublished archival source] BCARS, GR-0224  Box 21(ledger), Victoria, B.C. British Columbia, Government Agent (Ass’t  Gold Commissioner ) Mining Records, (1865-1867, 1887-1896), Lillooet, [unpublished archival source] BCARS, GR-0224  Box 22 (ledger), Victoria, B.C. British Columbia, Government Agent (Ass’t  Gold Commissioner ) Mining Records,  (1858-1872), Yale, [unpublished archival source] BCARS, GR-0252  Box 12  (ledger), Victoria, B.C.; also (in part), BRSC-ARC-1258 Box 24 file 1 UBC Library Special Collections, Vancouver B.C. British Columbia, Government Agent (Ass’t  Gold Commissioner ) Mining Records, (1858-1866), Yale, [unpublished archival source] BCARS, GR-0252  Box 12a  (ledger), Victoria, B.C. British Columbia, Government Agent (Ass’t  Gold Commissioner ), Manual of Record, (1873-1888), Lytton and Yale, [unpublished archival source] BCARS, GR-0252  Box 13  (ledger), Victoria, B.C. British Columbia, Government Agent (Ass’t  Gold Commissioner ), Manual of Record, (1880-1890), Lytton and Yale, [unpublished archival source] BCARS, GR-0252  Box 14 (ledger), Victoria, B.C. British Columbia, Government Agent (Ass’t  Gold Commissioner ) Mining Records, (1859-1874), Lytton Division, Yale District, [unpublished archival source] BCARS, GR-1054  Box 1, Folder 1 (ledger), Victoria, B.C. Brunsden D., (2001) “Back A’long: a Millennial Geomorphology” in Higgitt, D., Lee, E.M., Geomorpological Processes and Landscape Change: Britain in the Last 1000 Years. Blackwell. Butlin, R.A., (1993) Historical Geography: through the gates of space and time. Edward Arnold London. Canada Dept. of Energy, Mines and Resources, Canada Centre for Mapping. Canadian Topographic Maps 1:50,000 [map sires]. Various Editions. 1:50,000, sheets: 92-H/5, 92-H-6, 92-H/11, 92-H/14, 92-I/5, 92-I/4, 92-I/12, 92-I/13, 92-P/4, 92-O/1, 92-O/8, 92-O/9, 92-O/16, 93-A/12, 93-B/1, 93-B/8, 93-B/9, 93-B/15, 93-B/16.  Ottawa: Canada Centre for Mapping. Carmichael, H., Moore, C.W., (1930) Placer-Mining in British Columbia: with special reports on Atlin, Queen Charlotte, Cariboo, Quesnel, and Omineca mining divisions. Columbia Department of Mines Bulletin no. 2 Centre for Topographic Information, (2000) Canadian Digital Elevation Data Government of Canada, Natural Resources Canada, Earth Sciences Sector. Sherbrooke, Quebec, Canada. Online. <http: //www.geobase.ca/geobase/en/find.do?produit=cded  >. Church, M., (1980) “Records of recent geomorphological events” in Cullingford, R.A., Davidson, D.A., Lewin, J., (eds.)   Timescales in Geomorphology John Wiley & Sons Ltd. Church, M., (1983) “Pattern of instability in a wandering gravel bed channel.” Special Publications of 108 the International Association of Sedimentations v. 6 pp. 169-180. Church, M., (2010) “Sediment Management in Lower Fraser River: Criteria for a Sustainable Long- Term Plan for the Gravel-Bed Reach.” A report for Emergency Mangement BC, Flood Protection Program. Online. <http://www.pep.bc.ca/floods/sediment_docs/srategy_dr_church.pdf>. Church, M., Ham, D., Hamish, W., (2001) Gravel Management in Lower Fraser River,  report for The City of Chilliwack.  Online <http://www.geog.ubc.ca/fraserriver/reports/ report2001.pdf> Church, M., Ham, D., (2004) Atlas of the alluvial gravel-bed reach of Fraser River showing channel changes in the period 1912-1999  Church, M., Jones, D., (1982) “Channel bars in gravel-bed rivers” in Hey, R.D., Bathhurst, J.C., and Thorne, C.R., eds. Gravel-Bed Rivers John Wiley and Sons, Chichester, U.K., pp 291-338. Church, M., McLean, D., Wolcott, J., (1987) “River Bed Gravels: Sampling and Analysis.” In Thorne, C.R., Bathurst, J.C, and Hey, R.D., eds. Sediment Transport in Gravel-bed Rivers. , Chichester, p. 43-79. Clague, J., Luternauer, J., Monahan, P., Edwardson, K., (1998) Quaternary stratigraphy and evolution of the Fraser delta.  Bulletin-Geological Survey of Canada. Conroy, (1862?) [annotated by R.C. Harris?] “map of middle Fraser River tributaries,”  R.C. (Bob) Harris Fonds, [unpublished archival source] UBC Library, Special Collections, Vancouver, B.C. Cui, Y., Parker, G., Lisle, T.E., Goot J., Maria, E., Hansler-Ball, Pizzuto, J., Nicholas E., Reed, J.M, (2003) “Sediment pulses in mountain rivers : 1 Experiments.” Water Resources Researchvol. 39 no. 9 Cui ,Y., Parker, G., Pizzuto, J., Lisle, T.E., (2003b) “Sediment pulses in mountain rivers : 2 Comparison between experiments and numerical predictions.” Water Resources Research v. 39 no. 9 Dassmann, R.F., (1999) “Environmental Changes before and after the Gold Rush” In California History Vol. LXXVII No. 4: A Golden State: Mining and Economic Development in Gold Rush California pp 52-77 University of California Press. Dawson, G.M., (1889) G.S.C. Vol. III The Mineral Wealth of British Columbia.  The Geological Society of Canada. Douglass, I., (1990) “Sediment Transfer and Siltation” in Turner II, B.L., Clark, W.C., Kates, R.W., Richards, J.F., Mathews, J.T., Meyer, W.B. (eds), 1990 The Earth as Transformed by Human Action: Global and Regional Changes in the Biosphere over the past 300 years. Cambridge University Press. Epner G., (1862) Map of the Gold Regions in British Columbia [Reproduction]. Historical Map Society of British Columbia Facsimile No. 1. Morris Printing Co. Ltd. Victoria, BC. Etcheverry, B.A., (1915) Irrigation Practice and Engineering vol 1: Use of Irrigation Water and Irrigation Practice. McGraw Hill, New York. ESRI, (2008) ESRI ArcMap 9.3.  Software. ESRI, (2010) “Interactive TIN Editing” Online. <http://resources.esri.com/help/9.3/ArcGISDesktop/ 109 com/samples/3d_analyst/tin_editing/interactive_tin_edit_tools/55858c48-0a03-48b3-a14f- a86e121d9588.htm> accessed 24 June 2010. Evans, S.G., Savignay, K.W., (1994) “Landslides in the Vancouver-Fraser Valley-Wistler Region” in Monger, J.W.H., Geology and Geological Hazards of the Vancouver Region, Southwestern British Columbia, Geological Society of Canada Bullitin 481 pp. 251-286. Eyles, N., Kocsis, S.P., (1989) “Placer gold mining in Pleistocene glacial sediments of the Cariboo District, British Columbia, Canada 1958-1988” Geoscience Canada v. 15 no. 4 Ferguson, R.I., (1977) “Meander migration: equilibrium and change” In Gregory, K.J. ed. River Channel Changes Chichester: John Wiles pp 235-248. Fischer (2002) “High-resolution multiproxy climatic records from ice cores, tree-rings, corals and documentary sources using eigenvector techniques and maps; assessment of recovered signal and errors.” The Holocene vol. 12 iss. 4 pp 401-419. Ferguson, R.I., (2003) “The Missing dimension: effects of lateral variation on 1-D calculations of fluvial bedload transport.” Geomorphology. 2003 vol. 56 pp. 1-14. Galieos, R.M., (1970) Gold mining and its effects on landscapes of the Cariboo. Thesis (MA) Univeristy of Calgary. GeoBC (2011) “BC Geographic Names” [database]. Online <http://archive.ilmb.gov.bc.ca/bcnames/> Accessed 2/2011. GeoBC Web Map Services (2011) “BC_BC_BC_XC1M_BCALB_1995_2004” [digital air photo coverage] 1m resolution. Online WMS Server <http://openmaps.gov.bc.ca/imagex/ecw_wms.dll?> Accessed 2/2011. Germanoski, D., Harvey, M.D., (1993) “Asynchronous terrace development in degrading braided channels.” Physical Geography v. 14 pp. 16-38 Gilbert, G.K., (1917) Hydraulic Mine Debris in the Sierra Nevada. States Government Print Office. Goudie, A.S., (1977) Environmental Change. Oxford Clarendon Press (2nd edn, 1983). Goudie, A.S. (1995) The Changing Earth: Rates of Geomorphological Processes. Oxford: Blackwell. Goudie A.S., (2000) The Human Impact on the Natural Environment 5th ed. Gurnell, A.M., Peiry, J-L., Petts, G.E., (2003) “Using Historical Data in Fluvial Geomorphology” in Kondolf G. H., Piegay, H., (eds) Tools in Fluvial Geomorphology John Wiley and Sons. Haggen, E. A. (ed) (1923) Mining and Engineering Record Volume XXXVIII No. 1. Vancouver, BC. Hales, W.J., (2000) The Impact of Human Activity on Deltaic Sedimentation, Marshes of the Fraser River Delta, British Columbia.  Unpublished PhD thesis, University of British Columbia. Harley, J.B., (1982) “Historical geography and its evidence; reflections on modeling sources,” in Baker A.R., and Billinge M. (eds.), Period and Place: Research Methods in Historical Geography. Cambridge University Press. Cambridge. Haigh, M.J., (1978) Evolution of Slopes on Artificial Landforms – Blaenavon Hickin, E.J.,  (1995) “Hydraulic geometry and channel scour: Fraser River, B.C., Canada.” In Hickin, E.J., (ed), River Geomorphology, , Wiley, Chichester, pp. 155-167. 110 Higgit, D.L., Lee, M.E. (eds), (2001) Geomorphological Process and Landscape Change: Britain in the Last 1000 Years Blackwell, Oxford. Higgit, D.L., Lee, M.E., (2001) “Geomorphology for the Third Millennium” in  Higgit, D.L., Lee, M.E., (eds), Geomorphological Process and Landscape Change: Britain in the Last 1000 Years Blackwell, Oxford. Hoffman, D.f., Gabet, E.J., (2007) “Effects of sediment pulses on channel morphology in a gravel-ed river.” GSA Bulletin v. 119 no. 1-2 pp. 116-125. Hooke, J., (2003) “Coarse sediment connectivity in river channel systems: a conceptual framework and methodology.” Geomorphology v. 56 pp. 79-94. Hooke, J., Kain, R.J.P., (1982) Historical Change in the Physical Environment. Butterworth Scientific, London. Howay, F.W., (1926) Early History of the Fraser River Mines, Charles F. Banfield, Victoria. Howay, F.W., (1914) British Columbia From the Earliest Times to the Present Volume II the S.J. Clarke Publishing Company, Vancouver. Hungr, O., Evans, S.G., Hazzard, J., (1999) “Magnitude and frequency of rock falls and rock slides along the main transportation corridors of southwestern British Columbia” Canadian Geotechnical Journal vol. 36 no. 2 pp. 224-238. Jacobson , R.., (1960) “Closing Statement: Linguistics and Poetrics” in Seboek, T.A. (ed.),  Style and Language, Cambridge, Mass. James, L.A., (1989) “Sustained storage and transport of hydraulic gold mining sediment in the Bear River, California.” Annals of the Association of American Geographers v. 79 pp. 570-592. James, L.A., (1993) “Sustained reworking of hydraulic mining sediment in California: G.K. Gilbert’s sediment wave model reconsidered.” Zeitschrift für Geomorphologie, 88, suppl., pp. 49-66 James, L.A., (1997) “Channel incision on the lower American River, California from streamflow gauge records.” Water Resources Research v. 33 no. 3 pp. 485-490. James, L.A., (1999) “Time and the persistence of alluvium: River engineering, fluvial geomorphology, and mining sediment in California.” Geomorphology vol. 31 pp 265-290. James, L.A., (2004) “Decreasing Sediment yields in northern California: vestiges of hydraulic gold- mining and reservoir trapping.” In Sediment Transfer through the Fluvial System Proceedings of the Moscow Symposium. James, L.A., (2006) “Bed waves at the basin scale: implications for river management and restoration.” Earth Surface Processes and Landforms vol. 31 pp. 1692-1706 James, L.A., (2010) “Secular Sediment Waves, Channel Bed Waves, and Legacy Sediment” Geography Compass Vol. 4 no. 6 pp. 576-598. Jones, D.K., (2001), “The Evolution of Hillslope Processes” in Higgitt, D., Lee, E.M., (eds), Geomorpological Processes and Landscape Change: Britain in the Last 1000 Years. Blackwell. Jordan, R.P., (1994) Debris Flows in the Southern Coast Mountinas, British Columbia: Dynamic Behaviour and Physical Properties.  University of British Columbia Thesis (PhD). 111 Jung, M.A., (1999) “Capitalism Comes to the Diggings: From Gold-Rush Adventure to Corporate Enterprise.” In California History Vol. LXXVII No. 4: A Golden State: Mining and Economic Development in Gold Rush California 52-77 University of California Press. Keller, E.A., Melhorn, W.N., (1978) “Rhythmic spacing and origin of pools and riffles.”  Geological Society of America Bulletin V. 89 pp. 723-730. Kennedy M., (2009) “Fraser River Placer Mining Landscapes.” BC Studies no. 160 pp. 35-66. Kent, W., (1895) The Mechanical Engineer’s Pocket Book first edition. Wiley and Sons, New York. Krippendorff, K., (2004) Content Analysis: An Introduction to Its Methodology Sage Publications, Inc. Knighton, A.D., (1989) “River adjustments to changes in sediment load: the effects of tin mining on the Ringarooma River, Tasmania.” Earth Surface Processes and Landforms v. 14 pp. 333-359. Knighton, A.D., (1991) “Channel bed adjustment along mine-effected rivers of northeast Tasmania.” Geomorphology v. 4 pp. 205-219. Koshmann, A.H., Bergendahl, M.H., (1968) Principle Gold-Producing Districts of the United States Geolgical Survey Proffesional Paper 610. US Government Printing Office, Washington DC Lancaster, S.T., Hayes, S.K., Grant, G.E., (2001) “Modeling sediment and wood storage and dynamics in small mountainous watersheds.” In Dorava J.M., Montgomery D.R., Palcsak B., Fitzpatrick F. (eds.) Geomorphic Processes and Riverine Habitat, Water Science and Application v. 4 American Geophysical Union: Washington DC pp. 85-102. Leopold, L.B., Wolman, M.G., Miller, J.P., (1964) Fluvial Processes in Geomorphology. .H. Freeman. San Fransisco. Limbaugh, R.H., (1999) “Making Old Tools Work Better: Pragmatic Adaptation and Innovation in Gold-Rush Technology”  In California History Vol. LXXVII No. 4: A Golden State: Mining and Economic Development in Gold Rush California . 52-77 University of California Press.  Lisle, T.E. (1997) “Understanding the Role of Sediment Waves and Channel Conditions Over Time and Space.” In Sommarstrom, S.,(ed.) What is Watershed Stability? A review of the foundation concept of Dynamic Equilibrium in watershed management. Proceedings of the Sixth Biennial Watershed Management Conference. University of California Water Resources Center Report No. 92. Lisle, T.E., (2004) “Sediment Waves.” In Goudie, A.S. (ed) The Encyclopedia of Geomorphology volume 2., Routledge Ltd.  pp. 938. Lisle, T.E., Cui, Y., Parker, G., Pissuto, J.E., Dodd, A., (2001) “The dominance of dispersion in the evolution of bed material waves in gravel bed rivers.” Earth Surface Processes and Landforms 26 1409-1420. Little, B.J., (1992) Text-aided archaeology. Press, Boca, Raton, FL. 233 p. Madej, M.A., Ozaki, V., (1996) “Channel Response to Sediment Wave Propagation and Movement, Redwood Creek, California, USA.” Earth Surface Processes and Landforms v. 21 pp. 911-927. Mark, D.M., Church, M., (1977) “On the Misuse of Regression in Earth Science.” Mathmatical 112 Geology vol. 9 No. 1. Marron D.C., (1992) “Floodplain storage of mine tailings in the Belle Fourche river system: a sediment budget approach.” Earth Surface Processes and Landforms v. 17 pp. 675-685 Marsh, G.P., (1874) The Earth As Modified by Human Action: A New edition of man and Nature. Retrieved as a Project Gutenberg EBook, release date: July 2004. Online.  <http://manybooks.net/ titles/gpmarshetext04rthmh10.html> Marshall, D., (2000) Claiming the Land: Indians, Goldseekers,and the Rush to British Columbia, PhD diss., University of British Columbia. Meade, R.H., (1982) “Sources, sinks, and storage or river sediments in the Atlantic drainage of the United States.” Journal of Geology v. 90 pp. 235-252. Meade R.H., (1985) “Wavelike movement of bedload sediment East For River, Wyoming.” Environmental Geology and Water Science . 7 pp. 215-225 Meyer-Peter E., Müller, R. (1948) Formulae for Bedload Transport. Proceeding of 2nd IAHR Congress, Stockholmpp. 39-64. Miller, D.M., (1984) “Reducing Transformation Bias in Curve Fitting.” The American Statistician Vol. 38 no. 2 pp. 124-126. Nation, H.T., [draftsman] (1910) Sketch Map of the Lillooet and Clinton Mining Division of B.C., Bureau of Mines, Victoria, BC in British Columbia Dept. of Mines and Petroluem Resources 1910 Annual Report of the Minister of Mine for the Year Ending Dec. 31st 1910. The Department of Mines and Petroleum Resources, Victoria. National Air Photo Library, (1928), A289 40-90 [aerial photograph] 1:8000 Ottawa: Department of Energy Mines and Resources National Air Photo Library, (1928), A297 60-97 [aerial photograph] 1:8000 Ottawa: Department of Energy Mines and Resources National Air Photo Library, (1928),  A298 1-2 [aerial photograph] 1:8000 Ottawa: Department of Energy Mines and Resources Nicholas, A.P, Ashworth, P.J., Kirkby, P.J., Macklin, M.G., Murray, T., (1995) “Sediment slugs: large- scale fluctuations in fluvial sediment transport rates and storage volumes.” Progress in Physical Geography v. 19 no. 4 pp. 500-519. Nir, D., (1983) Man, a Geomorphological Agent: an Introduction to Antrhopic Geomorphology, Jerusalem: Keter. Noble, J.C., Tongway, D.J., (1986) “Herbivores in arid and semi-arid rangelands.” in Australian soils: the human impact J.S. Russell, RF Isbell (eds) University of Queensland Press in association with the Australian Society of Soil Science; Manchester, NH, USA. Northern Light, Whatcom (Steilacoom), Washington Territory, (1858), UBC Library, Koerner (microfilm), Vancouver B.C. Oreskes, N., Sharder-Frechetter K., Belitz K (1994) “Verification, validation and confirmation of numerical models in the earth sciences.” Science v. 263 pp. 641-646 113 Ormsby M.A., (1971) British Columbia: a History. MacMillan of Canada. Vancouver, BC. Otago Regional Council (2008) Channel Morphology and sedimentation in the Lower Clutha River. Online.<http://www.orc.govt.nz/Documents/Publications/Research%20And%20Technical/ Channel%20Morphology/Clutha%20River%20Channel%20Morphology%20web.pdf>  Accessed Sept. 2010. Petts G. (1989) “Historical Analysis of Fluvial Hydrosystems in Petts G., Holler H., and Roux, A., Historical Change of Large Alluvial Rivers: Western Europe. John Wiley & Sons. Petts G., Holler H., and Roux, A., (1989) Historical Change of Large Alluvial Rivers: Western Europe. John Wiley & Sons. PEP (Provincial Emergency Program) (2007) “Weekly Incident Situation Report” Period 29 Oct- 4 Nov. Emergency Management British Columbia. Online: <http://www.pep.bc.ca/incidents/ incidents-2007/Incidents-10_29-04.pdf> Accessed Nov. 2010. Pioneer and Democrat, Olympia, Washington Territory, 1858-1861, UBC Library, Koerner (microfilm), Vancouver B.C. Pitlick, J., (1993) “Response and recovery of a subalpine stream following a catastrophic flood.” Geological Society of America Bulletin . 105 pp. 657-670. The R Foundation for Statistical Computing, (2009) R 2.9.0 Software. Richards, K., (1982) Rivers: Form and Process in Alluvial Channels. Methuen. New York. Richards, K., (1984) “Book review: Hooke, J.M. and Kain, R.P., 1982: Historical change in the physical Environment” in Progress in Physical Geography  vol 8 p 147. Ricketts, A.H., (1943) American Mining Law with Forms and Precedents. 4th ed. enlarged and revised. California Division of Mines Bulletin 123, Feb. 1943. Roberts, N., (1998) The Holocene: an Environmental History. Oxford: Blackwell Roberts, R.G., Church, M., (1986) “The sediment budget in severely disturbed watersheds, Queen Charlotte Ranges, British Columbia.” Canada Journal Forestry Research v. 16 pp. 1092-1106 Rohe, R., (1986) “Man and the Land: Mining’s Impact in the Far West.” Arizona and the West vol. 28 pp. 299-338. Russell, S.O,. (1982) “Flood Probability Estimation.” Journal of the Hydraulics Division, Vol. 108, No. 1, pp. 63-73. Ryder, J.M., Church, M. (1986) “The Lillooet terraces of Fraser River: a palaeoenvironmental enquiry.” Canadian Journal of Earth Science vol. 23 pp. 869-884. Salmon J.H.M (1963) A history of goldmining [sic] in New Zealand. Re.Owen, Govt. Printer, Wellington, N.Z. Simmons, I.G., (1996) Changing the Face of the Earth: Culture, Environment, History, 2nd edn. Oxford: Blackwell. Smith, A.G., (1974) “Peak flows by the slope-area method.” Technical Bulletin no. 79, Inland Waters and Lands Directorate. Sutherland, D.G., Hansler M.E., Hilton, S.J., Lisle, T.E., (1998) “Sediment wave evolution and 114 channel morphologic changes resulting from a landslide, Navarro River, northwestern California.” Ecological Society of America Abstracts with Programs v. 30 iss. 7 p. A-360. Sutherland, D.G., Hansler Ball, M., Hilton, S.J., Lisle T.E., (2002) “Evolution of a landslide-induced sediment wave in the Navarro River, California.” GSA Bulletin v. 114 no 8. pp. 1036-1048. Tacconni, P., Laronne, J. B. and Duncan, M. J. (1992) “Bedload transport paths and gravel bar formation.” in Billi, P., Hey, R. D., Thorne, C. R. Tacconni, P. Dynamics of Gravel-bed Rivers, , Chichester.  pp. 177–200. Thomas W.L. (Ed.), (1956) Man’s Role in Changing the Face of the Earth. University Of Chicago Press. Trimble, S.W., (1983) “A sediment budget for Coon Creek basin in the Driftless area, Wisconsin, 1853-1977.” American Journal of Science v. 283 pp. 454-474 Trimble, S.W.,  (2008) “The use of Historical Data and Artifacts in Geomorphology” Progress in Physical Geography vol. 32 iss. 1 pp 3-29. Trimble S.W., and Cooke R.U., (1991) “Historical Sources for Geomorphological Research in the United States” Professional Geographer vol. 43 iss. 2 pp. 212-256. Turner, T.R., (1995) Geomorphic response of the Madison River to pint sediment loading at the Madison Slide, southwest Montana. MS thesis, Montana State University: Bozeman; 99 p. Turner II, B.L., Clark, W.C., Kates, R.W., Richards, J.F., Mathews, J.T., Meyer, W.B. (eds), (1990) The Earth as Transformed by Human Action: Global and Regional Changes in the Biosphere over the past 300 years. Cambridge University Press. Victoria Gazette (and Anglo-American)/Victoria Gazette, (1858-1859) Victoria British Columbia, UBC Library, Special Collections and UBC Library Koerner (microfilm) Vancouver , B.C. W. Oliver, (1862) The Handbook of British Columbia and Emigrant’s Guide to the Gold Fields, with Map and Two Illustrations. London. Waddington, A., (1858) The Fraser River Mines Vindicated, or, The History of Four Months, Victoria, P.DR. Garro. Ward, B., Harris, R.C., (2001a)  “Interior Fur Trade” Early Maps of British Columbia area: [cartographic material] a project listing all known maps, charts and reconnaissance sketches of early exploration and settlement during the pre-colonial and colonial eras to 1871. Vol. 6. [unpublished archival source] UBC Library Special Collections, Vancouver B.C. Ward, B., Harris, R.C., (2001b) “Cariboo” Early Maps of British Columbia area: [cartographic material] a project listing all known maps, charts and reconnaissance sketches of early exploration and settlement during the pre-colonial and colonial eras to 1871. Vol. 13. [unpublished archival source] UBC Library Special Collections, Vancouver B.C. Wathen, S.J., and Hoey, T.B., (1998) “Morphological Controls on the Downstream Passage of a Sediment Wave in a Gravel-Bed Stream.” Earth Surface Processes and Landforms v. 23 pp 715- 730 Wentworth, C.K., (1922) “A Scale of Grade and Class Terms for Clastic Sediments” The Journal of 115 Geology Vol. 30 pp. 377-392. Whitney, G.G., (1994) From coastal wilderness to fruited plain: a history of environmental change in temperate North America, 1500 to present. Cambridge University Press. Wilcock, P.R., Crowe, J.C., (2003)  “A surface-based transport model for sand and gravel”. Journal of Hydraulic Engineering. vol.129. no. 2 pp. 120 - 128 Wolcott, J., Church, M., (1991) “Strategies for sampling spatially heterogeneous phenomena; the example of river gravels.” Journal of Sedimentary Research . 61 no. 4 pp. 534-543. 116 -1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 -2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 represent the relative elevation of the current mine surface. Contours Colors represent the depth of sediment removed by mining, Black Circles 6 8 5 8 represent locations of point estimates of the depth of sediment removed by mining. Associated labels indicate the estimated depth. where color bands represent 2 m of removed material: where color bands represent 1 m of removed material: APPENDIX A: MINE SITE SURVEYS Mines are depicted with contours representing the surveyed surface.  For descriptions of the survey methodologies see chapter 4.  Color bands represent the amount of material removed or deposited relative to the interpolated original surface.  Because the planmetric areas of the mines are variable, maps are not depicted at a singe scale. Similarily, because the amount of relief on the surface of different mines is highly variable, the contour intervals for the current surface contours are not consistent between maps.  Also, the depth of material removal is highly variable, so the removed material contours (color bands) are not consistent between maps.  A set of map keys is provided on this page that can be matched to the appropriate maps by the text on those maps.  In all cases, orange represent places on the surface of the mine where the current surface is above the original surface and yellow-green through magenta represent places where material was removed.  The north direction for each map is towards the top of the page unless otherwise noted. 117 6 6 4 4 9 6 8 5 8 6 7 7 4 5 8.6 7.5 11 Mile Flat Hydraulic Depths based on point estimates No contours available Color bands represent 1 m of removed material ! ! ! ! ! ! ! ! ! ! 5 5 5 5 8 12 4.4 8.5 5.9 10.7 7 mi S. of Quesnell Hydraulic Depths based on point estimates No contours available Color bands represent 1 m of removed material 0 200100 Meters 0 200100 Meters W. Onion Bar Sluice Contour Interval 1m Color bands represent 1 m of removed material 0 10050 Meters    118 removed material. 0 200100 Meters 0 10050 Meters0 10050 Meters 119 1 k Above High Bar Contour interval: 1m. Color bands represent 1 m of removed material 1/2 k S. of Ward Creek Contour interval: 1m. Color bands represent 2 m of removed material W. High Bar GS Contour interval: 1m. Color bands represent 1 m of removed material W. High Bar Sluice Contour interval: 1m. Color bands represent 1 m of removed material ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! 6 6 6 684 6 4 46 6 66 6 4 4 5 5 4 4 6 6 6 6 8 8 8 6 3 6 6 3.9 0 10050 Meters 0 10050 Meters 0 10050 Meters 0 2010 Meters 0 15075 Meters High Bar NE Groundsluice Depths based on point estimate. No countours available. Color bands represent 1 m of removed material              120 Lower Leon Contour interval: 1m. Color bands represent 2 m of removed material Upper Leon Contour interval: 1m. Color bands represent 2 m of removed material 2 k. N. of Leon Contour interval: 2m. Color bands represent 2 m of removed material ¢ 0 200100 Meters 0 10050 Meters 0 10050 Meters                   121 Mines W of Skwish Creek Contour interval: 1m. Color bands represent 1 m of removed material. Bold outlines indicate areas where material was removed by mining and associated gullying 0 10050 Meters a) b) c) d) 122 E. Lee Sluice Sites Contour interval: 1m. Color bands represent 1 m of removed material    	  E. Lee Hydraulic Countour interval: 5m. Color bands represent 2 m of removed material 0 5025 Meters 0 10050 Meters Partial Survey of Fountain Hydraulic Countour interval: 2m. Color bands represent 1 m of removed material 0 5025 Meters        	       a) b) c) 123 Lilloet Mill Hydraulic Contour interval: 1m. Color bands represent 2 m of removed material Lillooet Old Bridge Hydraulic Partial Survey (site extends to the NE) Contour interval: 1m. Color bands represent 1 m of removed material 0 5025 Meters 0 9045 Meters Texas Creek Hydraulic Contour interval: 2m. Color bands represent 2 m of removed material Part of Foster's Bar Sluice Contour interval: 1m. Color bands represent 1 m of removed material 0 10050 Meters 0 5025 Meters              	        124  Selected Laluwissin Fan Sites Countor intervall: 1m. Color bands represent 1 m of removed material 0 10050 Meters 0 10050 Meters Siwhe Ground Sluice Contour interval: 1m. Color bands represent 1 m of removed material Cameron's Bar Sluice Contour interval: 1m. Color bands represent 1 m of removed material Surveyed with Tape and Level 0 10050 Meters Izeman North Sluice Contour interval: 1m. Color bands represent 1 m of removed material 0 10050 Meters                      a) b) c) 125 ! !! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! 5 0 0 0 0 0 0 0 0 20 20 15 20 3 3 32 33 2 2 3 3 86 3 3 1 0 2 1 1 2 5 6 4 4 3 4 1 2 4 1 4 1 2 4 3 0 3 6 6 5 3 4 8 7 4 6 7 3 2 3 4 4 4 4 4 3 6 6 4 4 4 2 4 6 6 64 3 3 3 2 5 2 3 5 2 2 2 2 4 6 5 4 8 8 0 0 0 4 4 3 3 3 6 2 2 6 6 6 0 0 0 1 4 2 4 9 1 4 8 24 4 3 3 3 6 11 1 3 10 16 10 10 18 10 Rip Van Winkle flat Depths based on point estimates No contours available Color bands represent 1 m of removed material 0 200100 Meters ! ! ! ! !! !!! ! ! ! ! !! ! ! ! ! ! 3 6 5 5 6 5 7 5 6 5 23 6 6 10 15 15 10 10 6.5 Splintnum Flat Groundsluice Depths based on point estimates No contours available Color bands represent 1 m of removed material 0 200100 Meters Hydraulic West of Rip Van Winkle Flat 126 3 3 2 5 3 6 1 1 3 2 4 3 5 5 3 5.5 1.75 3.75 0 200100 Meters 9 9 9 08 8 8 7 6 3 0 9 7 7 7 8 2 12 10 15 15 20 2017 12 12 20 20 12 20 12 12 17 15 15 20 18 18 18 1521 13 10 15 16 11 10 1021 17 1614 12 15 16 17 18 16 Emory Hydraulic Survey based on tape and clinometer cross sections and point estimates. No contours available. Point estimates are elevation above the river. Color bands represent 1 m of removed material 0 10050 Meters North Bend Depths based on point estimates No contours available Color bands represent 1 m of removed material 127 0 10050 Meters Trafalgar Sluice Survey based on tape and clinometer cross sections. Contour Interval: 2m. Color bands represent 1 m of removed material 0 200100 Meters American Creek Groundsluice Tape and clinometer cross sections. Contour Interval: 2m Color bands represent 1 m of removed material     128 APPENDIX B: GRAIN SIZE SAMPLING DATA Table B1: Wolman Samples Values presented are percent finer, by number for Wolman counts. For details of count methodology see Chapter 5.  Most samples presented in the following tables are lag deposits from mine sites.  Exceptions are the two RVW tailings fan samples, which were taken from a tailings fan below a hydraulic pit at Rip Van Winkle Flat and “Kersely Slide Bar” which was collected from the surface of a newly formed bar below the Kersely Slide (see chapter 7 for further discussion of this sample). 8 0 4 17 31 34 0 0 0 6 11.3 0 4 19 39 38 0 0 0 6 16 0 4 21 45 45 0 1 0 6 22.6 0 8 24 54 47 0 2 0 9 32 0 9 28 58 51 1 4 0 13 45.3 0 14 32 65 55 3 6 1 13 64 15 38 41 70 57 16 19 2 19 90.5 41 57 53 79 66 43 39 4 34 128 65 79 72 90 82 81 70 12 53 181 76 96 87 99 96 89 85 12 66 256 93 100 94 100 100 96 94 31 81 360 100 98 100 100 56 88 512 100 75 91 724 92 100 1024 95 2048 99 4096 100 n= 112 90 293 105 110 108 109 93 32 8 30 7 0 13 0 27 0 0 0 11.3 32 9 0 15 0 35 0 0 0 16 36 12 0 21 0 55 0 0 0 22.6 36 13 0 26 0 72 2 0 0 32 36 15 0 34 3 91 6 0 0 45.3 37 21 12 36 5 99 11 0 1 64 45 33 22 40 24 100 23 0 2 90.5 54 52 45 47 54 50 3 22 128 65 75 63 57 95 78 27 62 181 77 91 94 72 100 97 54 83 256 85 95 96 85 99 78 94 360 92 98 100 94 99 88 99 512 95 100 100 100 97 100 724 98 100 1024 99 2048 100 n= 92 129 49 286 102 113 102 100 155 size finer than (mm) W. Onion Bar Big Bar FerryW. Bank Big Bar Hydraulic ½ km S or Ward Ck, upper ½ km S of Ward Ck, lower Wheel- barrow Flat upper Wheel- barrow Flat lower Lower Leon W Skwish a) size finer than (mm) W Skwish b) c) e) Lillooet Old Bridge Hydraulic SIR5 Cobble SIR5 Back Channel Gravel Siwhe Sluice Stacked cobbles Siwhe Sluice above bulk Cameron's Bar 129 RVW* 11 RVW 8 RVW 9 RVW 7 RVW 6 RVW 5 RVW 3 8 0 0 0 0 0 2 0 0 0 11 1 0 0 0 0 2 0 0 0 16 1 0 0 0 0 4 0 0 0 23 1 0 3 0 2 8 0 7 3 32 1 0 3 0 2 10 0 7 3 45 4 0 6 0 3 10 4 11 3 64 11 10 12 3 6 28 6 20 5 91 30 23 33 23 16 50 22 53 16 128 76 72 52 39 37 64 41 76 27 181 91 90 70 52 49 72 51 80 43 256 98 95 85 65 67 80 67 91 54 360 100 100 94 87 83 92 90 98 76 512 97 97 98 98 96 98 95 724 100 100 100 100 100 100 100 n= 102 39 33 31 63 50 49 45 37 James 3 James 5 James 7 8 0 0 0 0 2 0 0 0 4 11 0 0 0 0 2 0 0 0 4 16 0 0 0 0 4 0 0 0 10 23 0 2 3 7 8 2 0 1 18 32 0 4 3 7 10 2 1 2 32 45 0 4 3 11 10 3 4 5 50 64 0 4 6 20 28 6 9 17 78 91 4 6 17 52 50 16 23 40 91 128 6 10 28 74 64 37 54 68 99 181 12 18 42 78 72 50 72 86 100 256 20 32 53 91 80 68 81 95 360 71 62 75 98 92 84 89 98 512 92 90 94 98 98 100 95 100 724 98 100 100 100 100 99 1024 100 99 2048 100 n= 49 50 36 46 50 62 99 94 117 size finer than (mm) Izeman N RVW3 pqn size finer than (mm) RVW Tailings fan top RVW tailings (in contact w river) James 3pqn Lytton Railroad Bride W North Bend Kersely Slide Bar * RVW denotes Rip Van Winkle Flat Table B2: Bulk Samples (on next page) Values presented are percent finer by weight.  For details of sampling methodology see Chapter 5. Most samples are unsorted sediment from mine scarps that probably is equivalent to the sediment that was removed by mining, except “Kersley Slide Bar,” which was taken from the surface of a newly formed bar in the active river channel below the Kersley Slide. 130 0.03 1.7 4.1 2.2 7.7 2.0 0.3 5.7 30.7 2.6 1.1 2.1 0.5 0.0 0.06 3.1 6.9 3.7 12.1 3.0 0.6 9.9 48.4 4.6 1.9 4.0 0.8 0.1 0.09 6.8 14.4 6.4 16.5 4.7 1.7 15.0 79.7 8.6 3.0 8.3 1.3 0.1 0.13 13.1 19.3 8.5 20.4 6.0 3.9 19.4 86.8 17.8 4.1 13.3 2.0 0.2 0.18 21.8 23.4 10.2 23.7 7.9 9.7 22.3 89.3 36.9 5.5 20.7 3.6 0.4 0.25 27.9 26.5 11.7 26.3 9.8 20.0 24.5 91.7 68.9 7.4 31.8 6.0 1.6 0.35 31.3 29.0 13.2 28.3 11.9 28.8 26.8 93.7 74.6 9.8 42.5 7.3 3.0 0.50 33.5 31.8 15.2 30.1 14.4 34.7 29.5 95.7 78.1 14.7 50.8 8.6 4.0 0.71 35.4 33.9 17.1 31.5 16.8 36.8 31.9 97.1 80.9 21.3 54.8 10.0 5.0 1 37.8 36.1 19.3 33.3 19.6 38.0 34.4 98.2 84.0 27.3 57.9 11.5 6.5 1.41 40.3 38.6 21.7 34.5 22.4 39.2 37.1 100 87.1 31.4 61.1 13.2 8.5 2 43.8 41.0 25.2 36.6 25.3 40.8 39.3 90.1 35.0 64.5 15.1 10.3 2.83 48.2 43.9 29.6 39.4 28.7 43.9 41.9 93.0 38.6 68.6 17.5 12.6 4 52.6 46.7 34.0 41.8 31.9 48.8 44.5 95.1 41.8 73.1 19.8 15.8 5.66 58.7 50.3 39.8 45.4 36.0 57.2 47.6 97.3 46.6 78.8 22.8 22.1 8 63.9 53.6 45.3 48.6 40.0 65.1 50.6 98.7 51.9 83.3 25.7 26.3 11.3 68.9 57.6 50.9 51.8 44.2 74.8 54.1 99.5 56.6 88.5 29.7 35.4 16 72.5 62.7 55.3 54.3 47.4 83.1 56.4 100 61.0 92.1 34.1 60.8 22.6 75.9 67.5 60.8 61.7 52.3 93.1 63.3 71.1 97.5 40.7 88.7 32 78.9 71.5 65.1 67.9 57.2 97.4 68.0 78.4 100 44.9 98.4 45.2 82.0 75.2 71.3 76.9 61.8 100 68.8 87.6 49.3 100 64 88.5 81.3 75.5 84.3 66.0 75.8 95.1 59.4 90.5 94.6 90.8 81.3 94.3 72.0 82.9 100 74.5 128 100 100 86.4 100 77.0 95.2 91.1 181 100 82.5 100 100 256 86.5 360 512 100 442.7 530.88 568.45 493.6 561.63 73.98 437.55 0.43 8.36 400.83 11.41 541.03 20.76 size finer than (mm) Big Bar Hydraulic ½ k south of Ward ck. (upper) Lower Leon W Skwish c) Lillooet Old Bridge Hydraulic Lillooet Mill Fosters Bar SIR5 Loess SIR5 Sand SIR5 Cobble SIR5 Back Channel Gravel Siwhe Sluice Kersely Slide Bar total sample weight (kg) Table B2: Bulk Samples (for description see previous page) 131 APPENDIX C: LIST OF ALL MAPPED MINES Tabel C.1: Location, calculated area, calculated volume, and which volume calculation method was used of all mapped mines.  “River Distance” is the distance from a mine’s outlet to Hope.  The value was calculated by placing nodes spaced 1 km apart along the river center line, finding the node that was closest to each mine’s centroid, and assigning that value to the mine. UTM coordinates are for the centroid of the mine polygon.  “Mine Type” is a description of the mine’s geometry.  See Chapters 2 and 3 for details describing the different mine types.  Abbreviations are as follows: S=sluice, GS=groundsluice, H=hydraulic, RB=rocker Box, M=Modern (post 1930s with tailings processed and dumped on site with no net addition of sediment to the river). Where a slash separates multiple abbreviations either both geometries are present at the mine site or it was impossible to differentiate between them.  They are listed in order of dominance/probability; for example, a GS/S site is dominated by groundsluice morphology but may have some area of parallel cobble stacks following classic “sluice” morphology.  “Volume Estimation Method” indicates how the volume was determined. Abbreviations are as follows: S=sluice regression, GS/H=groundsluice and hydraulic regression, DE=direct estimation of the volume by multiplying an estimated depth by the site area (see chapter 6 for further description of this methodology), Srv.=On Site Topographic Survey (see chapter 4) and NV means that a trivial net volume was dumped in the river by the mine and 0 volume was assumed.  Confidence is an indication of the researcher’s confidence that a given site is indeed a placer mine and of the accuracy of the polygon mapping and resulting area estimation.  1=high confidence, 2=medium high confidence, and 3=medium confidence.  See Chapter 3 for details.  The estimated volumes here are based the un-back transform corrected figures.  To get the proper back transform corrected figures, sluice site volumes should be multiplied by 1.31 and groundsluice and hydraulic site volumes should be multiplied by 1.23. 516 526544 5885030 H GS/H 191726 1,497,633 3 516 525348 5885534 S S 122260 550,858 3 514 526118 5883547 S S 16417 48,716 2 501 527821 5879725 S S 29941 100,677 2 498 530765 5879180 H GS/H 102409 720,770 3 497 530995 5877519 S S 35381 123,173 2 497 530909 5877880 S S 32310 110,377 2 496 531442 5877257 GS NV 59332 0 1 496 530802 5876983 S S 12607 35,411 3 496 531027 5877282 S S 15036 43,810 2 495 531006 5876337 M NV 6248 0 1 495 531320 5875957 S DE 46929 46,929 1 494 531759 5875199 GS GS/H 18790 99,767 1 494 531540 5875818 GS GS/H 6133 27,034 1 494 531266 5875062 S S 20228 62,689 1 494 531651 5875599 S S 3229 6,832 1 489 532634 5871675 S S 40245 143,910 1 488 535240 5871977 H GS/H 81446 551,817 3 River Distance (kmN of Hope) Position (UTMzone 10) Easting Northing Mine Type Volume Estimation Method Area (m2) Volume (m3) Mapping Confidence 132 487 533116 5869412 S S 20852 65,032 2 485 532735 5868035 D NV 92930 0 2 481 534353 5864760 S DE 472801 945,602 2 479 534336 5863471 S S 6736 16,608 1 479 534670 5863565 S S 65710 260,195 1 478 534682 5862487 S S 6366 15,512 3 478 535017 5862570 S S 48994 182,513 3 477 535611 5861663 H 107964 794,615 1 476 536001 5860710 GS GS/H 581 1,731 1 476 535937 5861156 H GS/H 2864 11,124 1 476 535947 5860853 RB NV 47711 0 1 476 536001 5861146 S S 17686 53,301 2 475 535974 5859933 GS GS/H 41316 250,065 1 470 537011 5855653 H 190415 1,207,231 1 467 536739 5852392 GS GS/H 35794 211,538 1 466 536120 5851502 S NV 75944 0 3 462 536832 5847849 S S 15900 46,869 2 461 535066 5846707 S S 19135 58,620 3 460 534755 5846581 H GS/H 8932 41,911 3 453 535066 5839530 GS GS/H 2513 9,551 1 453 535055 5839634 GS GS/H 4392 18,315 1 452 535279 5839115 GS GS/H 99710 698,665 1 447 535283 5834450 M NV 125101 0 1 446 534296 5834048 S S 50903 191,138 2 438 535288 5828222 S S 2439 4,868 2 435 535586 5826222 GS GS/H 8929 41,894 2 435 535918 5825499 H GS/H 5723 24,938 2 433 536078 5824017 S S 11143 30,506 3 431 535542 5822094 RB NV 35368 0 1 430 536033 5821748 S 25907 119,106 1 430 535702 5821887 S 26323 130,237 1 419 539385 5811770 S S 31222 105,903 3 416 540691 5809781 S S 44960 164,518 2 408 542227 5801823 H GS/H 3220 12,752 3 408 542218 5801927 H GS/H 3775 15,351 3 408 542270 5802078 H GS/H 6701 29,976 3 406 543457 5800639 S S 57871 223,179 1 394 549946 5792766 H GS/H 41038 248,104 3 392 550120 5791443 H GS/H 11846 58,255 3 392 550114 5791558 H GS/H 9324 44,064 3 390 550154 5789453 RB NV 16138 0 1 388 549510 5786882 S S 8676 22,547 1 388 549707 5787450 S S 1114 1,889 2 381 549841 5779967 H GS/H 5480 23,708 3 Table C.1 ctd River Distance (kmN of Hope) Position (UTMzone 10) Easting Northing Mine Type Volume Estimation Method Mapping Confidence Srv. Srv. Srv. Srv. Area (m2) Volume (m3) 133 366 549866 5765696 S S 36170 126,498 3 365 550191 5765130 S S 18629 56,753 1 364 550057 5764237 S S 40610 145,488 2 360 549895 5760221 S S 52808 199,812 3 359 549768 5758731 S S 5967 14,345 3 359 549739 5758889 S S 6826 16,876 1 359 549708 5759307 S S 1306 2,289 2 353 550126 5753277 S S 17766 53,593 1 341 541438 5747409 S/RB DE 71605 71,605 3 339 542124 5745389 S S 19508 60,004 1 333 540885 5740624 S S 10635 28,834 2 333 541021 5740880 S S 22397 70,897 2 326 542217 5734586 S S 12733 35,839 1 325 541816 5732862 H GS/H 743 2,306 1 325 541393 5732746 S S 22353 70,729 1 325 541592 5733037 S S 4929 11,388 1 325 542016 5733355 S S 11338 31,152 1 325 541714 5733376 S S 2599 5,256 1 324 541654 5731791 H GS/H 987 3,211 1 324 541650 5731991 H GS/H 1609 5,678 2 324 541710 5732654 H GS/H 763 2,379 1 324 541408 5731814 S S 5179 12,090 1 324 541670 5731828 S S 6607 16,224 1 324 541308 5732061 S S 14041 40,333 2 324 541322 5732389 S S 385 523 1 324 541677 5732458 S S 30845 104,360 1 324 541329 5732472 S S 2132 4,138 1 324 540589 5732515 S S 18972 58,018 1 323 541924 5731653 H GS/H 2859 11,101 2 321 543626 5730443 GS GS/H 100516 705,256 1 321 543153 5730600 GS GS/H 4010 16,471 1 321 543890 5730604 GS GS/H 1304 4,444 2 321 543712 5730616 GS GS/H 4370 18,208 2 321 543073 5730627 GS GS/H 3009 11,783 1 321 543545 5730642 GS GS/H 1390 4,787 2 321 543346 5730739 GS GS/H 3841 15,664 2 320 544185 5730111 GS GS/H 26879 151,466 2 320 544210 5730312 GS GS/H 2742 10,573 2 320 544205 5730458 GS GS/H 2934 11,441 3 319 544500 5728551 S S 32544 111,344 1 319 544205 5729221 S S 6999 17,394 1 316 544542 5726456 S S 1668 3,076 1 314 545047 5724781 H GS/H 41030 248,048 3 314 545929 5724192 S S 4679 10,694 1 Table C.1 ctd River Distance (kmN of Hope) Position (UTMzone 10) Easting Northing Mine Type Volume Estimation Method Mapping Confidence Area (m2) Volume (m3) 134 314 545882 5724208 S S 7594 19,196 1 314 545746 5724346 S S 243 300 1 313 545704 5723348 S S 4189 9,356 1 313 545792 5723509 S S 12146 33,853 1 313 546093 5723847 S S 11653 32,200 1 311 546644 5721816 RB NV 3936 0 3 311 546575 5721782 S S 34963 121,417 3 308 546911 5719694 S S 1000 1,658 3 307 547651 5718271 S S 14474 41,840 1 307 547531 5718496 S S 11522 31,763 1 305 548048 5716432 S S 1923 3,653 1 305 548120 5716490 S S 18086 54,761 1 305 547976 5716669 S S 3828 8,391 1 304 548215 5716179 S S 29298 98,071 1 303 548749 5714633 S S 6355 15,480 1 303 549078 5715103 S S 9164 24,088 2 302 548539 5714351 S S 51839 195,391 1 299 548960 5711167 S S 3351 7,145 1 299 549000 5711329 S S 1117 1,895 1 298 549163 5710506 S S 44836 163,970 1 296 550016 5708294 S S 339 449 1 296 549990 5708521 S S 7460 18,787 1 296 549886 5708635 S S 562 827 1 296 549843 5708684 S S 6538 16,020 1 296 549755 5708750 S S 1900 3,600 1 293 549694 5705455 S S 25338 82,292 1 292 550405 5704756 S S 26217 85,753 1 292 550157 5705106 S 57954 259,054 1 291 550710 5704106 H GS/H 27361 154,639 1 291 551016 5704073 S S 31286 106,165 1 290 550948 5703225 S S 4525 10,270 1 290 550889 5703622 S S 1583 2,888 1 289 551637 5702513 S S 32131 109,639 1 288 551569 5701684 S S 2559 5,159 2 285 553331 5699115 S S 5882 14,099 2 284 554266 5698541 S S 2452 4,899 2 283 554871 5697609 S S 5564 13,183 2 283 554919 5697938 S S 16672 49,632 2 282 554853 5697097 S S 5440 12,829 2 281 554333 5695992 S S 1165 1,994 1 278 553697 5692958 S S 1003 1,664 2 274 552620 5689219 S S 7234 18,102 2 270 552952 5685372 S S 1104 1,869 2 270 552956 5685836 S S 1126 1,914 1 Table C.1 ctd River Distance (kmN of Hope) Position (UTMzone 10) Easting Northing Mine Type Volume Estimation Method Mapping Confidence Srv. Area (m2) Volume (m3) 135 269 553264 5684886 S S 166 189 3 268 553269 5683681 S S 1696 3,139 1 266 554784 5682149 S S 341 452 1 266 554729 5682285 S S 2108 4,082 1 264 556115 5681471 GS GS/H 94630 657,332 1 263 556882 5680721 S S 3684 8,012 3 263 556607 5681060 S S 27352 90,257 1 262 556855 5679347 S S 7320 18,362 3 262 556991 5679910 S S 4596 10,465 1 261 556718 5678704 S S 151 169 2 261 556681 5678968 S S 11396 31,344 1 260 557422 5677670 S S 43266 157,060 1 260 557131 5678233 S S 3386 7,235 1 259 557540 5677449 GS GS/H 184 453 2 259 557499 5677465 GS GS/H 169 410 2 259 557488 5677486 GS GS/H 528 1,548 2 259 557776 5676706 S S 14360 41,442 1 258 557882 5676298 S S 1292 2,259 1 257 558441 5675218 S S 43323 157,310 1 257 558370 5675360 S S 871 1,403 1 257 558060 5675511 S S 3325 7,078 1 256 558724 5675282 S S 2050 3,946 1 253 559414 5672533 GS GS/H 1536 5,379 2 253 559379 5672776 GS GS/H 2632 10,080 2 252 559617 5671924 RB NV 37999 0 3 251 560497 5670494 S 7391 168,640 1 249 561631 5670302 H GS/H 81572 552,813 1 249 562201 5669917 H 143265 974,460 1 244 566143 5667835 S S 2109 4,084 1 244 565830 5667945 S S 1131 1,924 1 243 566361 5667203 S S 13168 37,323 1 241 566538 5665027 S S 31557 107,277 1 241 567142 5665338 S S 14945 43,490 1 240 566772 5664650 S S 57370 220,847 1 239 567471 5663941 GS GS/H 5048 21,543 1 239 567176 5663494 S 82718 222,285 1 238 567970 5662657 S 50514 70,176 1 237 568863 5662064 S S 1393 2,475 1 237 568864 5662362 S S 2362 4,683 1 236 569392 5661595 S S 3038 6,347 1 235 570658 5661391 GS GS/H 17440 91,458 1 235 570668 5661306 GS 22828 57,089 1 235 570339 5661496 S S 592 880 1 234 571629 5661057 GS GS/H 1704 6,071 1 Table C.1 ctd River Distance (kmN of Hope) Position (UTMzone 10) Easting Northing Mine Type Volume Estimation Method Mapping Confidence Srv. Srv. Srv. Srv. Srv. Area (m2) Volume (m3) 136 234 571588 5661088 GS GS/H 757 2,357 1 234 571566 5661122 GS GS/H 408 1,146 1 234 571184 5661252 GS 2344 4,032 1 234 570994 5661593 S S 39336 139,993 1 234 571262 5660919 S 25293 71,579 1 233 571801 5660859 GS GS/H 8067 37,217 1 233 571984 5660626 GS 97294 550,032 1 232 572147 5660202 S S 75127 305,887 1 230 572965 5658111 S S 23609 75,558 1 229 573543 5657827 S S 55830 213,706 1 228 573584 5656369 S S 11913 33,070 1 226 574073 5654574 S S 36021 125,869 1 225 574523 5653867 S S 3950 8,715 1 223 575609 5651808 H GS/H 47007 290,678 2 223 575185 5652664 H 39577 390,643 1 222 576394 5651204 S S 12293 34,348 1 221 575997 5650961 GS 183792 382,802 2 221 576312 5650255 S S 1813 3,402 1 221 576376 5650363 S S 1446 2,589 1 221 576500 5650627 S S 10490 28,359 1 221 576535 5650883 S S 3958 8,737 1 221 576469 5651030 S S 3639 7,893 1 221 576544 5651075 S S 4502 10,207 1 220 576314 5649475 S S 35746 124,709 1 220 576259 5650164 S S 937 1,533 1 217 577565 5647296 S S 240 296 1 215 579267 5645700 S S 4247 9,513 1 211 579174 5642535 S S 771 1,211 1 210 579024 5640894 S S 29652 99,504 1 209 579192 5640774 S S 26830 88,181 1 208 579445 5639819 S S 3662 7,954 1 208 579438 5639956 S S 29400 98,484 1 207 580001 5638245 GS GS/H 1582 5,567 1 207 579987 5638288 GS GS/H 2266 8,465 1 207 579902 5638327 GS GS/H 1951 7,109 1 207 579955 5638495 S S 3090 6,478 1 207 579863 5638250 S 8297 11,631 1 207 579825 5638439 S 1605 1,001 1 207 580162 5638509 S 13819 73,241 1 207 579790 5638614 S 1014 783 1 207 579885 5638487 GS 4632 4,156 1 205 579689 5636502 S S 5910 14,180 1 204 579181 5635469 S S 1590 2,903 1 204 579406 5635982 S S 5714 13,614 1 Table C.1 ctd River Distance (kmN of Hope) Position (UTMzone 10) Easting Northing Mine Type Volume Estimation Method Mapping Confidence Srv. Srv. Srv. Srv. Srv. Srv. Srv. Srv. Srv. Srv. Area (m2) Volume (m3) 137 203 578927 5634802 S S 2196 4,288 1 203 579141 5635231 S S 4407 9,948 1 200 579324 5632063 S S 2212 4,326 1 199 578833 5630776 S S 972 1,602 1 197 578924 5629658 GS GS/H 690 2,115 1 197 579346 5629633 H 45119 242,289 2 197 579106 5629473 S 2761 8,124 1 197 579005 5629504 S 5224 9,549 1 194 581187 5627094 H GS/H 81674 553,619 2 191 581254 5624904 S S 15487 45,403 1 190 581647 5623492 S DE 52046 5,205 1 188 580107 5622165 H GS/H 5446 23,536 1 188 579906 5622207 H GS/H 8839 41,402 1 188 580218 5622233 H GS/H 2659 10,201 1 188 580275 5622472 H 21468 106,696 1 188 580287 5622420 S DE 15360 15,360 1 187 579745 5621928 H GS/H 1405 4,848 1 187 579777 5622097 H GS/H 3506 14,083 1 187 579342 5621955 S S 8868 23,151 1 187 579067 5622040 S S 4198 9,381 1 186 578696 5622867 H GS/H 5116 21,882 1 186 578594 5623169 RB? DE 17635 17,635 1 186 578431 5622937 S S 2563 5,169 1 186 578407 5623074 S S 7878 20,066 1 185 578344 5623297 S S 18786 57,331 1 185 578221 5623300 S S 10646 28,870 1 185 578420 5623546 S S 4775 10,960 1 184 577459 5624169 S S 100620 435,354 1 181 575144 5622720 S S 18419 55,981 1 181 574743 5623206 S S 7577 19,144 1 181 575500 5623219 S S 8490 21,964 1 181 574991 5622951 S/RB DE 2408 2,408 1 181 575279 5622981 S/RB DE 105836 105,836 1 180 575223 5622207 GS GS/H 82418 559,505 2 176 576677 5618278 H GS/H 5648 24,558 1 176 576741 5618336 H GS/H 5196 22,281 1 176 576557 5618118 H DE 28797 155,504 1 176 576794 5617852 S S 7712 19,557 1 176 576475 5617962 S S 8684 22,572 1 176 576768 5618529 S S 19788 61,045 2 175 576255 5617719 H GS/H 28565 162,603 1 175 576398 5617944 H GS/H 4360 18,159 1 175 576342 5616966 S S 34637 120,051 1 175 576150 5617253 S S 67937 270,885 1 Table C.1 ctd River Distance (kmN of Hope) Position (UTMzone 10) Easting Northing Mine Type Volume Estimation Method Mapping Confidence Srv. Srv. Srv. Srv. Area (m2) Volume (m3) 138 175 576667 5617445 S S 17843 53,873 1 175 576363 5617853 S S 14436 41,707 1 174 575155 5616337 GS NV 33646 0 1 174 576095 5616361 H GS/H 8005 36,883 1 174 575577 5616320 S S 21522 67,565 1 173 576211 5615767 GS GS/H 24830 138,088 1 173 575925 5615954 H 11780 75,129 1 173 575505 5616032 S S 11230 30,793 2 172 575871 5614584 S S 33189 114,015 3 172 576030 5615292 S DE 231379 717,275 1 171 576685 5613943 GS GS/H 9356 44,240 1 171 576542 5614286 GS GS/H 9601 45,594 1 171 576688 5613794 S S 12129 33,796 1 171 576582 5614048 S S 12941 36,547 3 171 576640 5614051 S S 12240 34,170 1 170 576842 5613418 GS GS/H 8119 37,497 1 170 576853 5613521 S S 7411 18,639 1 155 584416 5601434 GS GS/H 35620 210,340 1 155 584482 5602210 S S 4631 10,562 1 154 584582 5600588 H GS/H 13483 67,747 1 154 584561 5600773 H GS/H 8212 37,998 1 154 584554 5601009 H 23341 207,204 1 153 585654 5599747 S DE 21458 21,458 1 151 586480 5598666 S S 122900 554,343 1 150 587386 5598060 S S 34951 121,367 1 149 588024 5597733 S S 974 1,606 1 148 588230 5597163 GS GS/H 576 1,714 1 148 588225 5597216 GS GS/H 520 1,521 1 148 588062 5597636 GS GS/H 1024 3,352 1 148 588090 5597510 S S 3606 7,807 1 148 588183 5597602 S S 41912 151,142 1 147 589251 5596420 S S 56432 216,492 2 147 588778 5596398 S DE 144485 144,485 1 146 589346 5596149 S S 62497 244,905 2 146 589730 5595335 S 51074 390,144 1 146 589406 5596221 S 15527 8,073 1 145 590110 5595528 S DE 32514 32,514 1 143 590584 5593183 S S 51451 193,626 1 143 590753 5593321 S S 60854 237,149 1 142 590914 5592617 S S 239076 1,238,430 1 139 592633 5589820 GS 1144 8,664 1 139 592564 5589871 GS 5066 24,791 1 139 592483 5589884 GS 4564 11,092 1 139 592343 5589857 S S 5267 12,338 1 Table C.1 ctd River Distance (kmN of Hope) Position (UTMzone 10) Easting Northing Mine Type Volume Estimation Method Mapping Confidence Srv. Srv. Srv. Srv. Srv. Srv. Srv. Area (m2) Volume (m3) 139 138 592702 5589793 GS GS/H 133 310 1 138 592555 5589481 S S 46120 169,660 2 136 593016 5587017 S S 21524 67,573 1 135 593259 5586224 GS GS/H 7526 34,322 1 135 593104 5586274 GS 22527 121,836 1 135 593603 5586568 S S 1143 1,949 1 135 593512 5586594 S S 4130 9,197 1 135 593154 5586639 S S 32012 109,149 1 133 593859 5584139 H GS/H 12046 59,403 1 132 593919 5583986 H GS/H 4046 16,643 1 132 593927 5584047 H GS/H 8812 41,255 1 132 594041 5583739 S S 14673 42,536 1 132 594331 5583903 S S 18245 55,343 1 132 594040 5583232 S 114183 773,019 1 131 594136 5582369 S 37704 70,239 1 130 594350 5581305 S S 37985 134,206 1 128 593902 5579737 S S 34193 118,194 1 125 595801 5577700 GS GS/H 5014 21,374 1 125 595584 5577459 S/GS 82290 554,039 1 124 595852 5576180 GS/S GS/H 39912 240,183 1 124 595304 5576684 S S 41700 150,219 1 123 595648 5575659 GS GS/H 5295 22,777 1 123 595768 5574892 H NV 23536 0 3 121 596465 5573935 GS GS/H 3296 13,104 1 120 596843 5573070 S S 40337 144,308 1 119 597532 5571596 S S 17898 54,074 1 118 598334 5571033 S S 32638 111,732 1 118 597901 5571092 S S 104076 453,481 1 117 598588 5570062 H GS/H 6606 29,481 2 117 598540 5570156 H GS/H 4909 20,853 2 117 598274 5570481 H GS/H 4915 20,882 2 117 598339 5570304 H 15262 208,856 1 116 599193 5568765 S 564927 1,651,433 1 115 599084 5568249 S S 18712 57,059 1 110 600972 5564194 S 37781 112,622 1 109 601433 5562938 S S 79837 329,202 1 108 601261 5562236 GS GS/H 51240 321,428 2 107 601211 5561680 GS GS/H 14523 73,879 2 107 601193 5561832 GS GS/H 10660 51,511 2 107 600768 5561339 H GS/H 44323 271,416 3 107 601261 5561088 H GS/H 37851 225,782 2 106 600898 5560694 GS GS/H 41607 252,120 3 105 601370 5559750 GS GS/H 1805 6,493 1 102 601430 5557030 H GS/H 3093 12,168 2 Table C.1 ctd River Distance (kmN of Hope) Position (UTMzone 10) Easting Northing Mine Type Volume Estimation Method Mapping Confidence Srv. Srv. Srv. Srv. Srv. Srv. Srv. Area (m2) Volume (m3) 140 99 602551 5554054 S S 11168 30,588 1 99 602341 5554122 S S 10847 29,529 2 99 602153 5554315 S S 620 931 1 99 602120 5554341 S S 2210 4,321 2 98 602908 5553086 S S 2039 3,921 3 98 602879 5553407 S S 4043 8,964 1 97 602524 5552068 S S 10794 29,355 2 97 602594 5552677 S S 7576 19,141 3 95 602589 5550563 RB NV 7820 0 1 95 602854 5550720 S S 14831 43,090 3 91 603654 5546987 S S 6802 16,804 1 91 603553 5547238 S S 13603 38,818 3 87 605113 5542862 S S 19337 59,369 1 85 605626 5540815 S S 13119 37,156 2 85 605882 5540935 S S 18476 56,190 2 85 605322 5541333 S S 21196 66,331 2 84 606154 5540488 S S 20820 64,912 3 83 606363 5539766 S S 18111 54,852 2 81 606744 5537079 S S 9995 26,751 1 76 609290 5534188 S S 35969 125,650 3 75 610065 5532707 GS GS/H 4123 17,013 2 73 610577 5531085 H GS/H 7167 32,421 2 73 610732 5530904 S S 1766 3,296 3 71 611222 5529416 H GS/H 4344 18,081 3 71 611556 5529638 H GS/H 8196 37,912 2 70 611208 5527921 H GS/H 126962 926,073 1 69 611577 5526748 H GS/H 6400 28,412 1 69 611502 5526899 H GS/H 2440 9,228 1 69 611410 5527036 H GS/H 27015 152,360 1 69 611721 5527093 H GS/H 6035 26,531 2 68 611490 5525736 H 207004 747,284 1 66 611835 5524270 H GS/H 30249 173,836 1 66 611689 5524365 H GS/H 6727 30,112 1 66 611349 5524381 H GS/H 2675 10,272 2 66 611323 5524452 H GS/H 2198 8,170 2 66 611320 5524495 H GS/H 934 3,011 2 66 611286 5524633 H GS/H 2763 10,668 2 65 611821 5523302 S S 121914 548,975 1 62 612688 5520381 S S 17360 52,117 3 61 612635 5519855 S S 5742 13,695 3 59 611207 5519132 S S 14278 41,156 3 55 611791 5515143 S S 8366 21,578 3 55 611514 5515528 S S 6458 15,783 3 48 613306 5509307 S S 6840 16,918 2 Table C.1 ctd River Distance (kmN of Hope) Position (UTMzone 10) Easting Northing Mine Type Volume Estimation Method Mapping Confidence Srv. Area (m2) Volume (m3) 141 47 613829 5508348 S S 20107 62,236 2 47 613662 5508829 S S 18221 55,255 2 46 614054 5507447 S S 6418 15,665 2 44 614702 5506131 S S 24762 80,037 2 44 614936 5506379 S S 20938 65,357 3 43 614633 5505257 S S 14915 43,385 1 43 614630 5505509 S S 21786 68,568 3 42 614547 5504601 S S 1188 2,042 2 41 614299 5503333 S S 17664 53,221 1 40 615105 5501856 GS GS/H 1286 4,372 1 40 615053 5501932 GS GS/H 600 1,797 1 40 615014 5501991 GS GS/H 482 1,392 2 40 614974 5502049 GS GS/H 571 1,696 2 40 614944 5502101 GS GS/H 585 1,745 2 38 615192 5500436 S S 13414 38,167 2 37 615307 5499776 S S 3958 8,737 2 36 616064 5498926 S S 64696 255,352 1 33 614716 5496627 S S 14708 42,658 1 30 615285 5494071 S S 13589 38,770 3 28 615752 5491919 S S 20382 63,266 2 24 613156 5490568 S S 209924 1,058,403 3 23 612702 5489911 S S 12545 35,201 3 22 613531 5488848 S/RB DE 76629 229,887 1 19 614580 5486043 S 168811 1,269,459 1 18 614670 5485192 S S 6453 15,769 2 16 614457 5482814 S S 49250 183,665 1 15 614572 5481935 S S 20944 65,379 1 14 614569 5481571 GS GS/H 6613 29,517 1 12 614551 5478917 GS GS/H 907 2,910 1 12 614561 5478974 GS GS/H 542 1,596 1 12 614566 5479011 GS GS/H 339 923 1 12 614537 5479086 GS GS/H 479 1,382 1 12 614536 5479236 GS GS/H 114 259 1 12 614511 5479511 GS GS/H 2978 11,642 1 12 614486 5479602 S S 1555 2,826 0 11 614516 5478407 GS GS/H 26850 151,276 2 11 614593 5478675 GS GS/H 29784 170,724 1 9 613530 5477187 GS GS/H 21184 114,742 1 9 613776 5476539 S S 3154 6,641 1 9 613807 5476923 S 35128 180,250 1 8 613677 5475670 S S 60881 237,276 3 6 614348 5474212 S S 95070 406,515 3 Table C.1 ctd River Distance (kmN of Hope) Position (UTMzone 10) Easting Northing Mine Type Volume Estimation Method Mapping Confidence Srv. Srv. Area (m2) Volume (m3) Map 2.1 Historical Minesites Along Fraser River From Hope to Quesnel Forks 

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}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            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.24.1-0071772/manifest

Comment

Related Items