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.

Item Metadata


24-ubc_2011_spring_nelson_andrew.pdf [ 29.01MB ]
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

Full Text

The Environmental History and Geomorphic Impact of 19 th 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  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 19 th 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 20 th 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.  ii  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.  iii  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  ................................ 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 Placer Mining Impacts on Geomorphic Systems  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  ................................................................. 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 and Cariboo Lakes  iv  ....................................................... 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 Geographic Variations in Methods  Constructing a Synthetic Grain Size Distribution to Represent All  ........................... 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 Material Dumped Into the River by Placer Mining  Sediment Transport Modeling to Estimate the Fraser and  ......................................................... 96 Virtual Velocity ............................................................................ 98 CHAPTER 8: CONCLUSION .................................................................. 102 Quesnel Rivers’ Capacities  v  WORKS CITED ............................................................................... 105 APPENDIX A: MINE SITE SURVEYS ................................................ 116 APPENDIX B: GRAIN SIZE SAMPLING DATA .................................... 128 APPENDIX C: LIST OF ALL MAPPED MINES ........................................ 131  vi  LIST  OF  TABLES  .................................................... 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 1.1: Key placer gold regions in British Columbia  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  ............................................................ 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 boxes, sluicing, and hydraulic mining  3.5: Example 95% confidence and prediction intervals spanning the range of mine site areas for both  .......................................... 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 the sluice and hydraulic and groundsluice regressions.  5.3: Calculated average annual bedload transport based on various input grain size distributions  ................................................................................ 98 5.4 Lengths of important canyons along the Fraser below Quesnel .................................... 99 at three locations.  vii  LIST OF FIGURES ...................... 6 2.1: General locations of administrative boundaries, region names, and some of communities ....... 19 1.1: Typical sluice box (A), ground sluice (B), and hydraulicked (C) placer sites  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  .......................................................................... 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 volume from mine area  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  ............................................................................................. 73 4.6: Cumulative distributions showing results of Wolman counts on mine lag deposits .......... 74 mines  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.  ..................................................  4.10: Cumulative distribution plots of representative grain size distributions for each facies. 4.11: Longitudinal distribution of gravel inputs to the Fraser  ....................................  75 76 80  4.12: Proportional symbol map of estimated mine excavation volumes along Fraser River between Hope and Cottonwood Canyon  .......................................................  81  viii  ...................................... 83 5.1: Tailings dump of the Bullion Pit Mine into the Quesnel River ............................. 85  4.13: Frequency magnitude distribution of mine site volumes  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  ..................................................................... 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 of the Fraser above Lillooet  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  ix  LIST  OF  MAPS  2.1: Historical Minesites Along Fraser River from Hope to Quesnel Forks  .....................  142  x  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.  xi  To “the creature” who will hopefully live in a world just as wondrous and inspiring as the one I know.  xii  CHAPTER 1: INTRODUCTION In the mid-late 19 th 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 20 th 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 19 th 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,  1  for 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 20 th 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 20 th 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  2  sediment 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 19 th 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  3  engineering 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 shortterm 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  4  between 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 20 th 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  5  Figure 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.  6  effects 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  7  because 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  8  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 goldbearing 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 20 th 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 20 th 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  9  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, treering, 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).  10  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)  11  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)  12  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  13  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 cutoff 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  14  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 19 th 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 mediumscale 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  15  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 19 th 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 19 th 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 19 th 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  16  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.  17  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  18  Early (~1859) Regional Names Couteau Country Canoe Country Cariboo Country  Vancouver  Mission  US A  Fra se r  r  ve Ri     Aggasiz    0  25  50km  rk Fo ll  Quesnel  io n s/ K  ei       D ivis    Quesnell Forks     Quesnellmouth Division      Williams Lake  ley        th    	   Big Bar            Lytton  T.  v. Di  Qu esn e  D  IS    v.    LI LL CA OO RI ET BO D O IST DI . ST .  Lillooet Fountian    Lillo oet D Cli nto i n     ET  Ash Yal cro e ft  LIL L O O  Boston Bar  v. Di iv. D  . YALE DIST  Yale    	   Hope              Barkerville  Figure 2.1 General locations of administrative boundaries, region names, and some of the communities referenced in the following text. Boundaries between these areas are somewhat fuzzy and variable 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. 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.  19  Table 2.1: Sources used in the creation of Map 2.1. Record Time Period or Year Published  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  Lytton and Yale  Manual of Record  GR-0252 Box 14  1880-1890  Lytton and Yale  Manual of Record  GR-1054 Box 1 Folder 1  1859-1874  Lytton  Mining Records  GR 0216 vol. 30-33  1860-1862  Alexandria  Mining Records  GR 0216 Vol. 76 Annual Reports to the Minister of Mines Conroy  1867-1899  Quesnellemouth  Mining Records  1874-1910  All  Mining Reports  1862  Map  2001 (a & b)  Middle Fraser Cariboo (including Quesnel River) All Lillooet and Clinton dvs. All  Map Collection  Various, Recent  All  Map Series  GeoBC  2011  All  Bancroft et al  1887  All  Geographic Names Database Published Narrative  Howay  1926  All  Published Narrative  Howay  1914  All  Published Narrative  Waddington  1858  Mid-Lower Fraser  Published Narrative  Victoria Gazatte  1858-1859  Mid-Lower Fraser  Newspaper  San Fransisco Evening Bulleting  1858-1859  Mid-Lower Fraser  Newspaper  Alta California (San Fransisco)  1858-1859  Mid-Lower Fraser  Newspaper  1858  Mid-Lower Fraser  Newspaper  Source  Bowman  1887  Epner  1862  Nation  1910  Ward and Harris Canadian Topographic Maps 1:50,000  Northern Light (Whatcom)  Map Map Map  20  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 20 th 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  21  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  22  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  23  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 mid1880s 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.  24  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  25  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).  26  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  27  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 20 th 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.”  28  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 185962. 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 20 thcentury, 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  29  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.  30  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.  31  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  32  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  33  wages amount of water  number of miners  duty of water  amount of gravel one miner can process duration of mining amount of gold concentration of gold  volume of gravel processed  Figure 2.2: Interrelation of proxies for placer mining gravel excavation volumes and practical and economic constraints. 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. 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. Rocker Box wages (dollars per day) 3/  amount of gravel one miner can process (yd day) 3  gold concentration (dollars/yd )  3  gold concentration (dollars/m ) generalized gold concentration (dollars/m3)  Hydraulic  --$3 (minimum)-$10 (good)-$50 (exceptional)-3  ½ yd  $6-$20-$100  value of gold gold concentration (oz/m3)  Sluice  10-15 yd3  highly dependent on conditions  24¢- 80¢-$4  3¢-$1 (10¢ normal)  --$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 10  0.18-0.61-3.1 0.6  0.02-0.08-0.76 0.08  34  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  35  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.  Year  Province Wide  1858 1859 1860  520,353 1,615,070 2,228,543  1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909  1,844,618 2,474,904 1,786,648 1,275,204 1,290,058 1,013,827 1,046,737 954,085 794,252 736,165 713,738 903,651 693,709 616,731 588,923 494,435 429,811 399,526 379,535 405,516 481,683 544,026 513,520 643,346 1,344,900 1,278,724 970,100 1,073,140 1,060,420 1,115,300 969,300 948,400 828,000 647,000 477,000  Quesnel/Keithley District & Lillooet Northern District Fraser in (excluding Swift and Cariboo Yale Division, Cayoosh Cottonwood Fraser River Creek) Division Rivers 520,353 538,744* 557135**  876,325‡ 676,325‡  200,000 995,082  10,000‡ 9,114 12,000 14,000 14,000 10,800 8,400* 6,000 5,000 15,000 29,000 25,000 20,000 14,500* 14,500* 9,000 6,400* 6,400* 3,800 3,800‡ 48,400 65,108 58,680 60,840 74,720 57,542 45,440 47,000 50,400 31,200 4,600 5,000 3,000 3,000 2,000  55,000 40,000‡ 25,000 32,455* 32,455* 39,910 81,800 40,717 64,200 68,000 107,934 94,700 120,000 73,750 37,660 26,239 41,455 52,506 39,763 51,376 30,267 40,663 33,665 37,480 42,614 42,700 36,905 26,080 27,440 25,820 34,500 30,000 16,800 12,000 13,200 10,000  38,084 40,716 95,905 12,833 50,910 162,700 95,300 207,500 100,860 92,740 101,200 71,000 72,800 79,520 80,127 57,500 54,150 41,300 44,500 37,050 46,700 86,910 197,050 200,000 214,860 193,300 510,000 240,000 160,000 132,000 150,000 96,000 39,600 44,000 30,000 12,000  12,000 30000† 48300† 22,630  11,650 4,860 4,860 11,400 7,000 5,700 6,520 8,200 8,300 6,750 4,500 4,500 2,000 36,000  * 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  36  Table 2.3b: Gold production values for regions in the Cariboo. Sources are the same as for table 2.2a. Quesnellemouth all Generall Fraser Quesnel/ Keithley below above Quesnel Quesnel Divison year 1858 1859 1860 1874 1875 1876 1877 1878 15,000 1879 62,700 1880 1881 82,300 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909  Main Stem of North Fork South Quesnel Fork Quesnel Quesnel Lower Upper [Cariboo] Quesnel Tribs.  200,000‡ 995,081‡ 4,603**  8,391 7,945* 7,500 5,000 6,500 7,000 10,700 16,120 17,000 17,000 17,500 12,000 10,000 7,500 7,000 6,000  19,616 74,354‡ 51,660 49,640 54,600 37,500 32,000 38,500 39,125 17,300 12,600 7,500 2,500 6,800 16,000 3,250  33,480** 40,716* 47952† 12833† 35910† 100,000  4930†  19719†  13446  32,693  11,000  8,200  10,400  8,600 5,400 5,400 5,000 3,500 4,900 5,402 5,200 5,500 5,000 7,000 4,000 6,000 7,360  6,000 8,400 8,000 8,000 6,000 5,000 6,000 5,000 3,750 4,000 5,500 6,500 4,000 3,000  19,100 16,000 15,000 9,000 8,600 5,000 2,600 3,000 3,000 2,500 8,500 3,750 3,500 3,000  8,000 8,300 9,000 4,500 12,000 10,000 10,000 10,000 11,200 10,000 9,500 8,500 9,000 63,500  †  †  9860  5,000  125,200  2,700  600 300 1,500 1,200 800  197,050 200,000 214,860 193,300 510,000 240,000 160,000 132,000 150,000 96,000 39,600 44,000 30,000 12,000  * 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.)  37  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. Estimated Gold Production (dollars) % hydraulic  % sluice  % rocker box  Estimated Gravel Moved (m3)  1,000,000 1,000,000 1,000,000 500,000 222,648 1,000,000 80,000 70,000 900,000 1,214,128 1,250,000 3,000,000 8,000,000 8,000,000  50 20 60 100 0 60 100 100 100 100 100 60 50 50  20 20 20 0 0 20 0 0 0 0 0 20 30 30  30 60 20 0 100 20 0 0 0 0 0 20 20 20  6,602,581 2,882,581 7,842,581 6,250,000 22,265 7,842,581 1,000,000 875,000 11,250,000 15,176,600 15,625,000 23,527,742 54,030,968 54,030,968  Regional Summary Values Quesnel River and Tributaries 9,236,776 Fraser River in Cariboo 2,000,000  76 35  11 20  13 45  89,411,768 9,485,161  Fraser River at Quesnel Fraser River, sundry bars in Cariboo Sundry Claims, N. Fork Quesnel Spanish Creek Golden River Quesnell South Fork -Sundry Claims Roses Gulch Chinese Farm Chinese Pit, Bullion Consolidated Cariboo (Bullion) Quesnel River Campan Creek Quesnel River, Sundry Claims Lightning Creek Tributaries of Willow River  Table 2.5: Estimates of the porportion of gold extracted in different regions and time perods by rocker boxes, sluicing, and hydraulic mining. Fraser River in Yale District RB H Sl  1858-59 1860's 1870's 1880's 1890's 1900's  50 40 40 15 5 50  50 55 55 25 5 50  0 5 5 60 90 0  Fraser River in Lillooet District RB H Sl  60 40 40 25 15 35  40 55 55 35 25 55  0 5 5 40 60 10  Fraser River near Quesnel RB H Sl  70 50 50 20 10 10  30 50 50 20 10 10  0 0 0 60 80 80  Quesnel River and tributaries RB H Sl  40 40 30 5 5  -----------40 40 30 5 5  20 20 40 90 90  Quesnel District (Fraser, Quesnel and tributaries) RB H Sl  70 45 45 25 10 10  30 40 40 30 5 5  0 15 15 45 85 85  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  38  39 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, Bridge River, and Cayoosh Creek) Quesnel District including Fraser River from Soda Creek to Cottonwood Canyon, and Quesnel River and tributaries, which for the time period from 1882-1890 is subdivided into Quesnell River and tributaries Fraser River from Soda Creek to Cottonwood Canyon  No Data from 1860-1874  Estimated Gravel Contribution (m3)  5,000,000  4,000,000  3,000,000  Pattern indicates the type of mining: Rocker Box Sluice Hydraulic (and some groundsluice)  2,000,000  1,000,000  0  0  0 19  90 18  80  74  18  18  60 18  58  18  Figure 2.3: Estimated total amount of gravel processed and the method of mining in several regions through time.  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 20 th 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: 1  In 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.  2  Gold 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  40  (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. Generously assuming an average mineable bar width of 60 m along this 26 km stretch of river yields  3  a total minable surface area only large enough to contain ~500 100 by 100 foot square claims. 4  For 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).  5  The 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.”  A miner’s inch is a measure of discharge equivalent to approximately 1.5 cubic feet (0.042 cubic  6  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).  41  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  42  Table 3.1: Landscape characteristics indicative of historical mining sites.  Sluice  Diagnostic  Supportive  Manually stacked cobbles and boulders  Linear heaps of tailings  Manually stacked cobbles and boulders Groundsluice Ditches leading directly to the scarp edge  Hydraulic  Remains of headboxes, high pressure iron pipe, hydraulic monitors  Necessary  Gully-like erosional pattern in locations where water would not be naturally concentrated  Teardrop shaped depressions at terrace edges with high scarps  Water source with high flow and substantial head (may be quite distant)  Ditches leading to or above the area  Presence of original terrace level “buttes” or barriers between an eroded pit and the downhill direction  All Clearly constructed drains leading through barriers  Remains of flumes, sluices  “ Unnatural” scarp patterns Plausibility of placer gold including sharp angles, presence (unless the site is freshness in areas where just of a prospect size) other erosion has not recently occurred, odd angles relative to nearby water features Nearby mining-era artifacts of human activity  Plausibility of access to a reasonable quantity (defined by the scale of the Soil and fines removed over site and technologies an area leaving unorganized used) of water coarse (cobble-boulder) lag deposits  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  43  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  44  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  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. Agency Date NAPL 1928 NAPL 1928 NAPL 1928 BC  1948  Roll A289 A297 A298  Photos 40-90 60-97 1-2  Region Yale Area Yale Area Yale Area  Scale 1:8000 1:8000 1:8000  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 N Bend to Lytton 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 various (~100 rolls, see list of BC 2005 works cited for complete list) Big Bar-Alexandria Area 1:30,000 BC 2006 Alexandria Area-Quesnel 1:30,000 (NAPL is the National Air Photo Library and BC is the British Columbia Base Mapping and Geomatic Services Branch.)  45  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.  46  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.  47  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  48  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.  49  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 500150,000 m2. At these sizes, manual surveys of individual sites were generally feasible and frequently  50  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.  51  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  52  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 premining 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 rasterbased 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.  53  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  54  Table 3.3 a): List of Mine Site Surveys Name  Location Easting Northing  Mine Type  unnamed  534681  5862484  sluice  7 mi S Quesnel  535406  5861808  hydraulic  11 Mile Flat  537029  5855632  sluice  Kersely Creek  536731  5852391  groundsluice  unnamed  535053  5839629  groundsluice  Bullion Pit  592209  5831550  hydraulic  Rocky Point West  535687  5821914  sluice  Rocky Point East  536004  5821762  sluice  W. Onion Bar  550156  5705121  sluice  Big Bar Ferry  560419  5670498  Big Bar Hydraulic  562184  1/2 km S of Ward Creek  Survey Type point depth estimates point depth estimates point depth estimates point depth estimates point depth measurements  Average Depth Original surface Area (m 2) Volume (m 3) (m) interpolation method 6366  25464  4  NA  107964  794615.04  7.36  NA  190415  1207231.1  6.34  NA  35794  137448.96  3.84  NA  4392  28328.4  6.45  NA  14634683.3  37.63  60% global polynomial interpolation  26468  119106  4.5  NA  29006  130236.94  4.49  NA  partial survey  57954  259054.38  4.47  TIN, manually edited  sluice/groundsluice  complete survey  28583  168639.7  5.9  75% global polynomial interpolation  5669938  hydraulic  complete survey  140210  974459.5  6.95  TIN, manually edited  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  sluice/groundsluice  complete survey  17512  57089.12  3.26  TIN, manually edited  W. High Bar GS  571183  5661257  groundsluice  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  sluice/ground sluice complex  98045  550032.45  5.61  NA  DEM from geobase 388910 point depth estimates point depth estimates  partial survey, extrapolated to whole site point depth estimates  55  Table 3.3 b): List of Mine Site Surveys Name  Location Easting Northing  2 k N of Leon  575195  5652688  Upper Leon  575272  5650910  Lower Leon  576026  5650808  579791  5638627  579827  Mine Type  Survey Type  hydraulic/ sluice  complete survey  ground sluice/hydraulic hydraulic/ground sluice  complete survey  Average Depth Original surface Area (m 2) Volume (m 3) (m) interpolation method TIN, minimal manual 56533 390643.03 6.91 editing global polynomial 19257 115734.57 6.01 interpolation  complete survey  56582  267067.04  4.72  sluice  complete survey  1283  782.63  0.61  5638445  sluice  complete survey  1589  1001.07  0.63  580191  5638384  sluice  point depth estimates  13819  73240.7  5.3  N/A  579901  5638327  eroded slope below mine  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  15307637.1  76.9  E Lee Hydraulic  579345  5629641  hydraulic  E lee a)  579181  E Lee b)  W Bank @ Skwish Creek a) W Bank @ Skwish Creek b) Skwish Creek W Bank @ Skwish Creek d) W Bank @ Skwish Creek c)  DEM from geobase 199059  TIN, manually edited 100% local polynomial interpolation 100% local polynomial interpolation  global polynomial interpolation global polynomial interpolation  complete survey  45119  242289.03  5.37  50% local 2nd order polynomial interpolation  5629603 ground sluice/chute  complete survey  1950  4348.5  2.23  TIN  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  Fountain Hydraulic  580265  5622481  hydraulic  partial survey  15360  106695  4.97  TIN, manually edited  56  Table 3.3 c): List of Mine Site Surveys Name  Location Easting Northing  Average Depth Original surface Area (m 2) Volume (m 3) (m) interpolation method  Mine Type  Survey Type  28797  155503.8  5.4  TIN, manually edited  Lillooet Old Bridge Hydraulic  576554  5618104  hydraulic  1/2 of site surveyed, average depth extrapolated to whole site  Lillooet Mill Hydraulic  575910  5615988  hydraulic  complete survey  12236  75129.04  6.14  TIN  Lillooet Mill  575956  5615565  sluice  total station point estimates  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  survey - limited extent  1928  8072.73  4.19  TIN, manually edited  SIR5  589687  5595398  sluice  complete survey  84556  339069.56  4.01  TIN, manually edited  Laluwissin e  592495  5589901  groundsluice/sluice  complete survey  2780  11092.2  3.99  Laluwissin d  592578  5589857  sluice/groundsluice  complete survey  4427  24791.2  5.6  Laluwissin c  592654  5589811  groundsluice/sluice  complete survey  1444  8664  6  Laluwissin a  592700  5589791  sluice/groundsluice  paced  96  576  6  NA  Siwhe Ground Sluice  593097  5586126  groundsluice  complete survey  39429  121835.61  3.09  TIN, manually edited  Cameron's Bar  594070  5583310  sluice  manual survey: 114183 tape, level and staff  773018.91  6.77  assumption of a flat plane  Izeman North  594155  5582433  sluice/groundsluice  complete survey  22369  70238.66  3.14  50% local polynomial interpolation  Splintlum Flat  595561  5577519  groundsluice/sluice  point depth estimates  82290  554039.11  6.73  NA  50% local polynomial interpolation & manual TIN creation global polynomial interpolation global polynomial interpolation  57  Table 3.3 d) : List of Mine Site Surveys Name Hydraulic N. of Rip Van Winkle  Location Easting Northing  Mine Type  598285  5570296  hydraulic  599108  5569023  sluice, limited hydraulic  598882  5568720  hydraulic  600999  5564186  sluice  North Bend  611463  5525745  sluice & hydraulic  Emory Flat  614656  5485726  American Creek  613541  Trafalgar  613884  Rip Van Winkle Flat hydraulic west of Van Winkle Flat Lytton Railroad Bridge W.  22008  208855.92  9.49  TIN  439211  1651433.36  3.76  NA  13438  98366.16  7.32  NA  23463  112622.4  4.8  TIN, manually edited  point estimates  207004  747284.44  3.61  NA  Sluice  3 cross sections, point estimates  168811  1269458.72  7.52  NA  5477207  sluice  4 cross sections  14968  80976.88  5.41  NA  5477066  sluice  8 cross sections  50209  180250.31  3.59  NA  Table 3.4: Average geometry of mines by technology. Depth (m) Area (m2) Volume (m3)  Survey Type Lytton Moran Survey point depth estimates point depth estimates 95% survey, 5% visual estimation  Average Depth Original surface Area (m 2) Volume (m 3) (m) interpolation method  Sluice  Groundsluice  Hydraulic  3.77 64,840 289,736  4.71 18,587 97,682  6.57 44,270 295,292  58                 Frequency     Frequency                                     Sluice: log(area)              Frequency    Frequency               Hydraulic and Groundsluice: log(area)            Hydraulic and Groundsluice: log(volume)                Sluice: log(volume)  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.  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 premining 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  59  2 k N. of Leon  ~100m  Big Bar Hydraulic  ~200m  1/2 k S of Ward  ~100m  Siwhe Groundsluice  ~100m  Figure 3.2: Rendered views of some selected mine site surveys (left column of images) and reconstructed pre-mining surfaces used to estimate the depth of excavations (right column). Scales are approximate because images are oblique perspective views.  60  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 premining 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.  61  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 lognormally 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%.  62        Sluice Regression    Log(volume)  Groundsluice and Hydraulic Regression    Symbols Represent Mine Type:    Hydraulic Groundsluice Sluice                Log(area) 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.  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  63      Residuals                Residuals                Fitted Log(area) Hydraulic and Groundsluice Regression            Fitted Log(area) Sluice Regression  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(volume m3)  Log(volume m3 )                Hydraulic Groundsluice        2  Log (area m )              2  Log (area m )  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.  64  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)  (ŷ|xh)±1.721√(0.00143+0.00397(xh-4.18) 2).  (3.4)  and for the groundsluice it is  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.  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. Predicted  Mine Area (m2) 1,000 10,000 15,136 21,380 100,000 1,000,000  Volume (m3) 1,658 26,767  Sluice CI +% -% 21 26 11 12  PI +% 81 75  Hydraulic and Groundsluice CI PI Volume (m3) +% -% +% -% 73 34 115 3,261 42 62 16 96 17 47,811 94 77,526 16 61 15 Predicted  -% 45 43  75 10 43 91 58 26 22 701,035 76 43 14 85 46 10,278,980 24 47 36 96 58 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.  67,025 432,116 6,975,896  11 16 32  65  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 This particular sequence is somewhat common. It possibly represents distinct phases in the mining  1  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 20 th 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.  66  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.  67  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  68  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 volumeby-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.  69  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     Legend Loess  Cobble Gravel  Sand  Boulder Gravel   	  Debris Flow   Figure 4.1: Example mine stratigraphies. Note the loess cap and variable composition of sediment that was removed by mining.  70      Depth of Facies (m)     Loess  Sand  Debris Flow  Fluvial Gravel  Figure 4.2: Box and whisker plots of the depth of each facies. 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  71     $            	          
         #  !" !" "           $      $-&* .&4" )) 	    	  	  +./"-. -  &((+  1 "(  +"/  &((  * ./"-&.'  #/"- /%" *)" +#  )&*" .),(" &*!& /". /%/ /%" .),(" 2.  +),+.&/" !",/% &*/"$-/"! .),(" 2&/% +".. &(0/&+* +# /%"  &".    +   ("  &*" &/". +2"- -."&((++"/ "-- ". 0".*"(  	  &2 %"       &(    (+"  / 3! -  0(&        &*"-    
  ' % **  "(    	  Figure 4.3: Cumulative distribution plots plots of all bulk mine samples by facies.        	  $-&* .&4" ,.&  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.  72  	      	   $-&* .&4" ))             # -!  " +*    -  '  3 !   * ./"-&.'  #/"- /%" *)" +#  )&*" .),(" &*!& /". /%/ /%" .),(" 2.  +),+.&/" !",/% &*/"$-/"! .),(" 2&/% +".. &(0/&+* +# /%"  &".   &*"-   +      "-&. (+2+((01&( +-!*   (01&( -1"(        &.% '2             $-&* .&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. 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  73  ' $ ( . ##        	    $'              ' $ ( . &(   Figure 4.6: Cumulative distributions showing results of Wolman counts on mine lag deposits. 7 Debris Flow and Colluvial  n=24  6 Fluvial Gravel  number of sites  5 4 3 2 1 0 91-128 64-91  180-256 512-1024 2048-4096 128-180 256-512 1024-2048  max lag grainsize (mm)  Figure 4.7: Histogram of the maximum grain sizes observed for all sites where Wolman counts were done.  74  	    !*#' +#1 && 	    	  	      ( %  (%# %#'+ * -%$ +&)%+ +" %#'+ * (%%&' (-',+ #* +&)%+ * ," +& (%(*  	  " #/         #% %( , 0 *-  %#       #'*    $ "   
  ''  %   *. %  (+,*+ *        	  !*#' +#1 )+#    lin  e          1:1    maximum grain size observed in lag deposit    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  Figure 4.9: Comparison of the largest stone observed in Wolman counts of lag deposits with the largest stone in corrosponding bulk samples.  75    	    (!% )!- $$ 	 	  	  	  Measured Values    (!) #&,&##*+!# &(% 	 #*+!# (+#          !%(  
  Estimated Values (!) #&,&##*+!# #*+!# (+#        	  (!% )!- ' ! 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). 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  76  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 Table 4.1: Percent by weight of grains in each facies that belong to various grain size classes. Fluvial Gravel Estimated Coarsest Finest boulders cobbles gravel granules and sand silt and smaller  0±7 34 ± 12 33 ± 9 29 ± 10 4±3  14 37 32 17 1  0 12 49 37 2  Debris Flow Estimated Coarsest Finest 2±2 22 ± 3 37 ± 7 27 ± 12 12 ± 7  0 29 23 45 3  0 23 38 33 7  Sand  Loess  0 0 7 88 5  0 0 0 52 48  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.  77  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 0.52 m) of the mine. This was done in order to produce a conservative estimate of the total volume  78  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.  79  3,500,000    	  3,000,000    gravel input/ 10 km of river m3    2,500,000  	  	  2,000,000    
   1,500,000  	   1,000,000 	   500,000  0  0  100  200  300  400  500  river distance upstream of Hope (km)  Figure 4.11: Longitudinal distribution of gravel inputs to the Fraser. Values are summed for 10km long computation cells.  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)  80  Cotto         Barkerville    sn    el     Quesnel  Q ue       River ood nw  	
   Ri  ve  r  il c  
   Ch     Quesnell Forks  ot in  R  ive  Williams Lake  r  Big Bar  Bridge  ve r Ri  ve Ri et oo  Li ll  r           Lillooet    Lytton  Boston Bar  Vancouver  Yale  Fras er Ri v  er  Mission Aggasiz US A   Hope        Figure 4.12: Proportional symbol map of estimated mine excavation volumes along Fraser River between Hope and Cottonwood Canyon.  81  Table 4.2: Various estimates of sediment excavation associated with placer mining. Estimated total sediment contribution based on gold production values from: Time Series Sum (1858-1860 & Haggen's (1923) 1874-1909) Estimate  Physically observed along the Fraser  Cottonwood River  700,000***  NA  NA  Lightning Creek‡  NA  54,000,000  NA  Quesnel River and Tributaries  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  (Values are in cubic meters.)  * **  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  ***  For the years 1872-1878 & 1882-1895  ‡  A tributary of Cottonwood R. Lillooet Division includes Bridge River and Cayoosh Creek Yale Division includes Thompson River  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,  82  nearly all of the hydraulic mining was probably done in the eighties through the first decade of the 20 th 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  120  Line Represents Cumulative Volume of Sediment Bars Represent Number of Mines 	  100   
	 	 80  60  60  40  Number of Mines  % Cumulative Volume of Sediment  80  40  20 20  0 100  1,000  10,000  100,000  1,000,000  0 10,000,000  Mine Volume m 3  Figure 4.13: Magnitude-frequency distribution of mine site volumes.  83  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 : The large difference between the largest stones encountered during bulk sampling and Wolman  1  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.  84  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.  85  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  86  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 1  Observed (Wolman) GSD Predicted GSD of Mobile Sediment  T  ve lF ac  in  ra uv ia l d= Fl gs  rd  pu t  Inp  ut  0.4  New  GS  D= Or  ch a  In  % finer  0.5  bar b elow Ke  0.6  tailings  I  rsely Slid e  G  u np  Eroded  S tG  M D=  fan belo w Van W inkle Fla  e  0.7  ie s  0.8  t  Smallest Flood at Marguerite Largest Flood at Hope Vertical Bars Represent the gs lin selective mobility threshold. ai  Ba r  0.9  0.3  0.2  0.1  0 -2  0  2  4  6  8  10  grainsize (psi) Figure 5.3: Cumulative grain size distribution plots of Wolman samples from an eroded and censored 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.  87  French Bar Active Tailings Dump  French Bar  Relict Tailings Fan  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.  88  Table 5.1: Hydraulic conditions at various flood flows at the Hope and Marguerite Gauging Stations. Average Average Shear Flow Depth Velocity Stress Froude (cms) (m) (m s-1) (pascals) Number Margeurite (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 mean annual flood 8666 10.26 ten-year flood 10800 11.24 largest annual flood 15200 13.24  3.95 5.63 6.41 7.65  46.69 55.38 60.63 71.44  0.43 0.56 0.61 0.67  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  89  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.  90  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 19 th 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 aggradationdegradation 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.  91  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 A) Dispersing Slug  A) Translating Slug  C) Translating and Dispersing Slug  Figure 5.5: Visual definitons of translation and dispersion.  92  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 and 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),  93  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  94  Table 5.2: Input grain size distributions for sediment transport calculations. Finer than (mm) 0.25 0.5 1 2 4 8 16 32 64 128 256 512  Orchard Bar 1 2 2 2 4 9 11 8 10 30 11 10  Fluvial Mine Gravel Tailings Quesnel --------(% fraction)------11 31 3 5 6 3 7 7 4 6 6 4 8 8 5 7 7 6 8 8 9 8 8 13 14 13 22 16 8 25 10 7  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).  95  16000  Fraser at Hope Fraser at Margurite Quesnel Near Quesnel  14000  discharge (cms)  12000 10000 8000 6000 4000 2000  0  50  100  150  200  250  300  350  average no. days per year of equivalent or greater discharge  Figure 5.6: Partial duration series for flows on the Fraser and Quesnel rivers.  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  96  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 12  Fraser at Marguerite  height above lowest point in cross section (m)  10 8 6 4  2 0 14  Fraser at Hope  12 10 8 6 4 2 0 4  Quesnel Near Quesnel  2 0 0  50  100  150  200  250  300  350  400  horizontal position (m)  Figure 5.7 Cross sections at locations of sediment transport calculations. See text for description of data sources.  97  Table 5.3: Calculated average annual bedload transport based on various input grain size distributions at three locations. Units are bulk cubic meters. Input Grain Size ----all sediment---Distribution Margurite Quesnel Hope Mine Tailings 29,000 8,885,000 14,324,000 Fluvial Gravel 3,087 3,368,000 6,061,000 Quesnel 80 Orchard Bar 297,000 696,000  Quesnel 2,890 551 17 -  ----gravel only---Margurite Hope 2,052,000 3,593,000 1,218,943 2,330,000 197,044 396,000  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  98  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 Table 5.4 Lengths of important canyons along the Fraser below Quesnel. Canyon btw. Yale and Spuzzum Black Canyon blw Hells Gate Hells Gate Above Kanaka Great Falls (Lillooet) Fountain Just above Fountain Pavilion (White Canyon) High Bar French Bar Churn Creek Iron Canyon English Bluff Chimney Creek to Wms. Lake Creek Soda Creek Canyon  Length (km) 1.3 2.8 2.7 4.1 1.0 4.5 0.8 6.3 1.0 1.2 13.5 1.4 2.2 11.9 1.6  99  ""' $( 
 ($) ! #"% # !	  " &! $%$  	 !)$  	 !)$  4    !)$  '%"  	 !)$  	 !)$   !)$  1  '$ " !  #%$( $&#"  0.5  0 1900  2000  2100  2200  2300  2400  $ 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). 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.  100  Endnotes Mr. Baerg is owner of Fraser River Raft Expeditions. He frequently leads expedition-length raft  1  trips down the river.  101  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). Historicaldocumentary 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  102  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 20 th 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).  103  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).  104  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 01570158, 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, 15BCC04040/00920094, 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  105  Base Mapping and Geomatic Services Branch (2005), 30BCC05006/0036-0037, 30BCC05006/00980099, 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, 30BCC05033/00430044, 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, 30BCC05054/00750076, 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, 30BCC05075/01960197, 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),  106  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  107  the International Association of Sedimentations v. 6 pp. 169-180. Church, M., (2010) “Sediment Management in Lower Fraser River: Criteria for a Sustainable LongTerm 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/  108  com/samples/3d_analyst/tin_editing/interactive_tin_edit_tools/55858c48-0a03-48b3-a14fa86e121d9588.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.  109  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 goldmining 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).  110  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  111  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: largescale 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  112  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 Oct4 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  113  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 715730 Wentworth, C.K., (1922) “A Scale of Grade and Class Terms for Clastic Sediments” The Journal of  114  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.  115  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.  Black Circles  Contours represent the relative elevation of the current mine surface.  8  8 6  5  represent locations of point estimates of the depth of sediment removed by mining. Associated labels indicate the estimated depth.  Colors represent the depth of sediment removed by mining, where color bands represent 2 m of removed material:  -2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32  where color bands represent 1 m of removed material:  -1 0 1 2 3 5 4 6 7 8 9 10 11 12 13 14 15 16 116  0  50  100 Meters  11 Mile Flat Hydraulic Depths based on point estimates No contours available Color bands represent 1 m of removed material 7      7.5   W. Onion Bar Sluice Contour Interval 1m Color bands represent 1 m of removed material  5  8.6  4 5.9 !  8.5 !  7  7 mi S. of Quesnell Hydraulic Depths based on point estimates No contours available 5 Color bands represent 1 m ! of removed material  6  10.7 !  8 5 !  8 6  5 6  9 4 5  12  5  !  4  !  !  4.4 ! 0  100  6  200 Meters  8  0  100  200 Meters  !  117  0  100  200 Meters  removed material.  0  50  100 Meters  0  50  100 Meters  118  0  50  100 Meters  0 10 20 Meters            1 k Above High Bar Contour interval: 1m. Color bands represent 1 m of removed material  W. High Bar GS 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            0  50  100 Meters    4 !  5 !5  4  6! 4  ! 6!4 6 !! ! ! 6! 4 6 ! 4! ! 6!  ! 6! !  8 4 ! ! ! 6  6  !  0  75  150 Meters  6 High Bar NE Groundsluice Depths based on point estimate. No countours available. Color bands represent 1 m of removed material  6 !  ! 6 !  3  6 ! 8 !  ! !8 8 3.9 ! ! 6 6 ! 6 ! ! !6  W. High Bar Sluice Contour interval: 1m. Color bands represent 1 m of removed material 0  50  100 Meters  119  Lower Leon Contour interval: 1m. Color bands represent 2 m of removed material                   0  100  200 Meters  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  50  100 Meters  0  50  100 Meters  120  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  a)  0  50  100 Meters  b)  d)  c)  121  Partial Survey of Fountain Hydraulic Countour interval: 2m. Color bands represent 1 m of removed material  E. Lee Hydraulic Countour interval: 5m. Color bands represent 2 m of removed material       	    0  25    50 Meters       0  50  100 Meters  E. Lee Sluice Sites Contour interval: 1m. Color bands represent 1 m of removed material  	     a)         c) b)  0  25  50 Meters  122  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  25  50 Meters  0  25    50 Meters  Part of Foster's Bar Sluice Contour interval: 1m. Color bands represent 1 m of removed material  Texas Creek Hydraulic Contour interval: 2m. Color bands represent 2 m of removed material  	               0  45  90 Meters  0  50  100 Meters  123  Selected Laluwissin Fan Sites Countor intervall: 1m. Color bands represent 1 m of removed material  c) b)             a)           0  50  100 Meters    Cameron's Bar Sluice Contour interval: 1m. Color bands represent 1 m of removed material Surveyed with Tape and Level    Siwhe Ground Sluice Contour interval: 1m. Color bands represent 1 m of removed material  Izeman North Sluice Contour interval: 1m. Color bands represent 1 m of removed material                  0  50  100 Meters  0  50  100 Meters  0  50 100 Meters  124  !  2 ! 4! 3 4 0 ! ! 1 10 ! 1 ! 3 6 4 ! ! ! 6 ! 5 ! 2 !  Rip Van Winkle flat Depths based on point estimates No contours available Color bands represent 1 m of removed material  7  !  3!  4!  1 !  8!  3!  !  !  3!  0  0 ! 15 !0 20 ! ! 20 0 5!  3  0  2!  !  !  3 ! 0 ! 20 0 ! 3 ! ! 3 ! !  !  6 8  2  16 4 !! 6 4 ! 6  4 ! ! 6 6  3 ! !  !  3 !  100  200 Meters  ! !  6 !5 7 ! 5 6 5 ! !  5 !  10 3 2  ! !!  6.5 6 !!  6 ! 0  100  200 Meters  0  3  18 !  0 3  !  2 !  4 0 ! 4 !  4  5 !  !  !  6 !  !  0  3 !  6  !  ! ! Hydraulic West of Rip Van Winkle Flat  10  4  4 !  1!  15!  4  6 ! ! 4 3 ! ! 4 ! ! ! 4 6 4 ! 2 ! ! 4 ! 4 6 ! 3 ! ! 5 6 3 ! ! ! 5 ! 3 5 ! 10 ! !2 3 1 !2 10 ! ! !2 ! !2 2 ! ! 2 ! 2 1 ! ! ! 4 8 2 5 ! !8 ! ! 6 ! !4 !0 ! 6 4 !0 0 0 ! ! ! !  !  !  !  4 !  ! !  6  7!  2! 4  Splintnum Flat Groundsluice Depths based on point estimates No contours available 10 Color bands represent 15! 1 m of removed material !  1  !  10 9 8! ! ! 4 1! 4 ! ! 4! 2 ! ! 3 3 36 !! !! 3 3! 3 ! 1 1 1 ! ! ! ! 2 3 ! !3 3 ! 2! 2  0 !  !  !  125  North Bend Depths based on point estimates No contours available 2 Color bands represent 1 m of removed material  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  3 5 17 14 16  5.5  9  7  7  8  12 15  7  16  5  17 18 2 16 15  15  15  15 21  2  15 11 10 8 16 10  8  7  3  3.75  6  3  0  18  10 18  18  3 4  1.75 12  1  12  17  3  9  20  20  12 17  20  1  20  12  6  21 15 13  10  8 0  20 12  5 9  3  3 20  0  100  200 Meters  0  50  12 9  100 Meters  126  American Creek Groundsluice Tape and clinometer cross sections. Contour Interval: 2m Color bands represent 1 m of removed material  Trafalgar Sluice Survey based on tape and clinometer cross sections. Contour Interval: 2m. Color bands represent 1 m of removed material      0  50  100 Meters  0  100    200 Meters  127  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). WheelWheel½ km S or ½ km S Big Bar barrow size finer W. Onion Ferry W. Big Bar Ward Ck, of Ward barrow Bank Hydraulic upper Ck, lower Flat upper Flat lower than (mm) Bar  8 11.3 16 22.6 32 45.3 64 90.5 128 181 256 360 512 724 1024 2048 4096 n=  size finer than (mm)  Lower Leon  W Skwish a)  6 6 6 9 13 13 19 34 53 66 81 88 91 100  0 0 0 0 0 0 15 41 65 76 93 100  4 4 4 8 9 14 38 57 79 96 100  17 19 21 24 28 32 41 53 72 87 94 98 100  31 39 45 54 58 65 70 79 90 99 100  34 38 45 47 51 55 57 66 82 96 100  0 0 0 0 1 3 16 43 81 89 96 100  0 0 1 2 4 6 19 39 70 85 94 100  112  90  293  105  110  108  109  0 0 0 0 0 1 2 4 12 12 31 56 75 92 95 99 100 93  SIR5 Back Channel Gravel  Siwhe Sluice Stacked cobbles  Siwhe Sluice above bulk  b)  W Skwish c)  e)  Lillooet Old Bridge SIR5 Hydraulic Cobble  8 11.3 16 22.6 32 45.3 64 90.5 128 181 256 360 512 724 1024 2048  30 32 36 36 36 37 45 54 65 77 85 92 95 98 99 100  7 9 12 13 15 21 33 52 75 91 95 98 100  0 0 0 0 0 12 22 45 63 94 96 100  13 15 21 26 34 36 40 47 57 72 85 94 100  0 0 0 0 3 5 24 54 95 100  27 35 55 72 91 99 100  0 0 0 2 6 11 23 50 78 97 99 99 100  0 0 0 0 0 0 0 3 27 54 78 88 97 100  n=  92  129  49  286  102  113  102  100  32  Cameron's Bar  0 0 0 0 0 1 2 22 62 83 94 99 100  155  128  size finer than (mm) Izeman N RVW* 11  8 11 16 23 32 45 64 91 128 181 256 360 512 724  0 1 1 1 1 4 11 30 76 91 98 100  n=  102  RVW size finer Tailings than (mm) fan top  8 11 16 23 32 45 64 91 128 181 256 360 512 724 1024 2048  0 0 0 0 0 0 0 4 6 12 20 71 92 98 100  n=  49  RVW 8  RVW 9  RVW 7  RVW 6  RVW 5  RVW 3  RVW3 pqn  0 0 0 3 3 6 12 33 52 70 85 94 97 100  0 0 0 0 0 0 3 23 39 52 65 87 97 100  0 0 0 2 2 3 6 16 37 49 67 83 98 100  2 2 4 8 10 10 28 50 64 72 80 92 98 100  0 0 0 0 0 4 6 22 41 51 67 90 96 100  0 0 0 7 7 11 20 53 76 80 91 98 98 100  0 0 0 3 3 3 5 16 27 43 54 76 95 100  33  31  63  50  49  45  37  North Bend  Kersely Slide Bar  0 0 0 0 0 0 10 23 72 90 95 100  39  RVW tailings (in contact w river)  James 3pqn  0 0 0 2 4 4 4 6 10 18 32 62 90 100  0 0 0 3 3 3 6 17 28 42 53 75 94 100  0 0 0 7 7 11 20 52 74 78 91 98 98 100  2 2 4 8 10 10 28 50 64 72 80 92 98 100  0 0 0 2 2 3 6 16 37 50 68 84 100  0 0 0 0 1 4 9 23 54 72 81 89 95 99 99 100  0 0 0 1 2 5 17 40 68 86 95 98 100  4 4 10 18 32 50 78 91 99 100  36  46  50  62  99  94  117  50  Lytton Railroad James 3 James 5 James 7 Bride W  * 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.  129  Table B2: Bulk Samples (for description see previous page)  size finer than (mm)  0.03 0.06 0.09 0.13 0.18 0.25 0.35 0.50 0.71 1 1.41 2 2.83 4 5.66 8 11.3 16 22.6 32 45.2 64 90.5 128 181 256 360 512 total sample weight (kg)  ½ k south of Ward Big Bar ck. Hydraulic (upper)  1.7 3.1 6.8 13.1 21.8 27.9 31.3 33.5 35.4 37.8 40.3 43.8 48.2 52.6 58.7 63.9 68.9 72.5 75.9 78.9 82.0 88.5 94.6 100  4.1 6.9 14.4 19.3 23.4 26.5 29.0 31.8 33.9 36.1 38.6 41.0 43.9 46.7 50.3 53.6 57.6 62.7 67.5 71.5 75.2 81.3 90.8 100  Lower Leon  2.2 3.7 6.4 8.5 10.2 11.7 13.2 15.2 17.1 19.3 21.7 25.2 29.6 34.0 39.8 45.3 50.9 55.3 60.8 65.1 71.3 75.5 81.3 86.4 100  W Lillooet Old Skwish Bridge Lillooet c) Hydraulic Mill  7.7 12.1 16.5 20.4 23.7 26.3 28.3 30.1 31.5 33.3 34.5 36.6 39.4 41.8 45.4 48.6 51.8 54.3 61.7 67.9 76.9 84.3 94.3 100  2.0 3.0 4.7 6.0 7.9 9.8 11.9 14.4 16.8 19.6 22.4 25.3 28.7 31.9 36.0 40.0 44.2 47.4 52.3 57.2 61.8 66.0 72.0 77.0 82.5 86.5  Fosters Bar  SIR5 Loess  SIR5 Sand  SIR5 Cobble  0.3 0.6 1.7 3.9 9.7 20.0 28.8 34.7 36.8 38.0 39.2 40.8 43.9 48.8 57.2 65.1 74.8 83.1 93.1 97.4 100  5.7 9.9 15.0 19.4 22.3 24.5 26.8 29.5 31.9 34.4 37.1 39.3 41.9 44.5 47.6 50.6 54.1 56.4 63.3 68.0 68.8 75.8 82.9 95.2 100  30.7 48.4 79.7 86.8 89.3 91.7 93.7 95.7 97.1 98.2 100  2.6 4.6 8.6 17.8 36.9 68.9 74.6 78.1 80.9 84.0 87.1 90.1 93.0 95.1 97.3 98.7 99.5 100  1.1 1.9 3.0 4.1 5.5 7.4 9.8 14.7 21.3 27.3 31.4 35.0 38.6 41.8 46.6 51.9 56.6 61.0 71.1 78.4 87.6 95.1 100  73.98  437.55  0.43  8.36  400.83  SIR5 Back Channel Gravel  Siwhe Sluice  Kersely Slide Bar  2.1 4.0 8.3 13.3 20.7 31.8 42.5 50.8 54.8 57.9 61.1 64.5 68.6 73.1 78.8 83.3 88.5 92.1 97.5 100  0.5 0.8 1.3 2.0 3.6 6.0 7.3 8.6 10.0 11.5 13.2 15.1 17.5 19.8 22.8 25.7 29.7 34.1 40.7 44.9 49.3 59.4 74.5 91.1 100  0.0 0.1 0.1 0.2 0.4 1.6 3.0 4.0 5.0 6.5 8.5 10.3 12.6 15.8 22.1 26.3 35.4 60.8 88.7 98.4 100  11.41  541.03  20.76  100 442.7  530.88  568.45  493.6  561.63  130  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. River Distance (km N of Hope)  516 516 514 501 498 497 497 496 496 496 495 495 494 494 494 494 489 488  Position (UTM zone 10) Easting Northing  526544 525348 526118 527821 530765 530995 530909 531442 530802 531027 531006 531320 531759 531540 531266 531651 532634 535240  5885030 5885534 5883547 5879725 5879180 5877519 5877880 5877257 5876983 5877282 5876337 5875957 5875199 5875818 5875062 5875599 5871675 5871977  Mine Type  Volume Estimation Method  Area (m2 )  Volume (m3 )  Mapping Confidence  H S S S H S S GS S S M S GS GS S S S H  GS/H S S S GS/H S S NV S S NV DE GS/H GS/H S S S GS/H  191726 122260 16417 29941 102409 35381 32310 59332 12607 15036 6248 46929 18790 6133 20228 3229 40245 81446  1,497,633 550,858 48,716 100,677 720,770 123,173 110,377 0 35,411 43,810 0 46,929 99,767 27,034 62,689 6,832 143,910 551,817  3 3 2 2 3 2 2 1 3 2 1 1 1 1 1 1 1 3  131  Table C.1 ctd River Distance (km N of Hope)  487 485 481 479 479 478 478 477 476 476 476 476 475 470 467 466 462 461 460 453 453 452 447 446 438 435 435 433 431 430 430 419 416 408 408 408 406 394 392 392 390 388 388 381  Position (UTM zone 10) Easting Northing  533116 532735 534353 534336 534670 534682 535017 535611 536001 535937 535947 536001 535974 537011 536739 536120 536832 535066 534755 535066 535055 535279 535283 534296 535288 535586 535918 536078 535542 536033 535702 539385 540691 542227 542218 542270 543457 549946 550120 550114 550154 549510 549707 549841  5869412 5868035 5864760 5863471 5863565 5862487 5862570 5861663 5860710 5861156 5860853 5861146 5859933 5855653 5852392 5851502 5847849 5846707 5846581 5839530 5839634 5839115 5834450 5834048 5828222 5826222 5825499 5824017 5822094 5821748 5821887 5811770 5809781 5801823 5801927 5802078 5800639 5792766 5791443 5791558 5789453 5786882 5787450 5779967  Mine Type  Volume Estimation Method  Area (m2 )  Volume (m3 )  Mapping Confidence  S D S S S S S H GS H RB S GS H GS S S S H GS GS GS M S S GS H S RB S S S S H H H S H H H RB S S H  S NV DE S S S S Srv. GS/H GS/H NV S GS/H Srv. GS/H NV S S GS/H GS/H GS/H GS/H NV S S GS/H GS/H S NV Srv. Srv. S S GS/H GS/H GS/H S GS/H GS/H GS/H NV S S GS/H  20852 92930 472801 6736 65710 6366 48994 107964 581 2864 47711 17686 41316 190415 35794 75944 15900 19135 8932 2513 4392 99710 125101 50903 2439 8929 5723 11143 35368 25907 26323 31222 44960 3220 3775 6701 57871 41038 11846 9324 16138 8676 1114 5480  65,032 0 945,602 16,608 260,195 15,512 182,513 794,615 1,731 11,124 0 53,301 250,065 1,207,231 211,538 0 46,869 58,620 41,911 9,551 18,315 698,665 0 191,138 4,868 41,894 24,938 30,506 0 119,106 130,237 105,903 164,518 12,752 15,351 29,976 223,179 248,104 58,255 44,064 0 22,547 1,889 23,708  2 2 2 1 1 3 3 1 1 1 1 2 1 1 1 3 2 3 3 1 1 1 1 2 2 2 2 3 1 1 1 3 2 3 3 3 1 3 3 3 1 1 2 3  132  Table C.1 ctd River Distance (km N of Hope)  366 365 364 360 359 359 359 353 341 339 333 333 326 325 325 325 325 325 324 324 324 324 324 324 324 324 324 324 323 321 321 321 321 321 321 321 320 320 320 319 319 316 314 314  Position (UTM zone 10) Easting Northing  Mine Type  Volume Estimation Method  Area (m2 )  Volume (m3 )  Mapping Confidence  549866 550191 550057 549895 549768 549739 549708 550126 541438 542124 540885 541021 542217 541816 541393 541592 542016 541714 541654 541650 541710 541408 541670 541308 541322 541677 541329 540589 541924 543626 543153 543890 543712 543073 543545 543346 544185 544210 544205 544500 544205 544542 545047 545929  S S S S S S S S S/RB S S S S H S S S S H H H S S S S S S S H GS GS GS GS GS GS GS GS GS GS S S S H S  S S S S S S S S DE S S S S GS/H S S S S GS/H GS/H GS/H S S S S S S S GS/H GS/H GS/H GS/H GS/H GS/H GS/H GS/H GS/H GS/H GS/H S S S GS/H S  36170 18629 40610 52808 5967 6826 1306 17766 71605 19508 10635 22397 12733 743 22353 4929 11338 2599 987 1609 763 5179 6607 14041 385 30845 2132 18972 2859 100516 4010 1304 4370 3009 1390 3841 26879 2742 2934 32544 6999 1668 41030 4679  126,498 56,753 145,488 199,812 14,345 16,876 2,289 53,593 71,605 60,004 28,834 70,897 35,839 2,306 70,729 11,388 31,152 5,256 3,211 5,678 2,379 12,090 16,224 40,333 523 104,360 4,138 58,018 11,101 705,256 16,471 4,444 18,208 11,783 4,787 15,664 151,466 10,573 11,441 111,344 17,394 3,076 248,048 10,694  3 1 2 3 3 1 2 1 3 1 2 2 1 1 1 1 1 1 1 2 1 1 1 2 1 1 1 1 2 1 1 2 2 1 2 2 2 2 3 1 1 1 3 1  5765696 5765130 5764237 5760221 5758731 5758889 5759307 5753277 5747409 5745389 5740624 5740880 5734586 5732862 5732746 5733037 5733355 5733376 5731791 5731991 5732654 5731814 5731828 5732061 5732389 5732458 5732472 5732515 5731653 5730443 5730600 5730604 5730616 5730627 5730642 5730739 5730111 5730312 5730458 5728551 5729221 5726456 5724781 5724192  133  Table C.1 ctd River Distance (km N of Hope)  314 314 313 313 313 311 311 308 307 307 305 305 305 304 303 303 302 299 299 298 296 296 296 296 296 293 292 292 291 291 290 290 289 288 285 284 283 283 282 281 278 274 270 270  Position (UTM zone 10) Easting Northing  545882 545746 545704 545792 546093 546644 546575 546911 547651 547531 548048 548120 547976 548215 548749 549078 548539 548960 549000 549163 550016 549990 549886 549843 549755 549694 550405 550157 550710 551016 550948 550889 551637 551569 553331 554266 554871 554919 554853 554333 553697 552620 552952 552956  5724208 5724346 5723348 5723509 5723847 5721816 5721782 5719694 5718271 5718496 5716432 5716490 5716669 5716179 5714633 5715103 5714351 5711167 5711329 5710506 5708294 5708521 5708635 5708684 5708750 5705455 5704756 5705106 5704106 5704073 5703225 5703622 5702513 5701684 5699115 5698541 5697609 5697938 5697097 5695992 5692958 5689219 5685372 5685836  Mine Type  Volume Estimation Method  Area (m2 )  Volume (m3 )  Mapping Confidence  S S S S S RB S S S S S S S S S S S S S S S S S S S S S S H S S S S S S S S S S S S S S S  S S S S S NV S S S S S S S S S S S S S S S S S S S S S Srv. GS/H S S S S S S S S S S S S S S S  7594 243 4189 12146 11653 3936 34963 1000 14474 11522 1923 18086 3828 29298 6355 9164 51839 3351 1117 44836 339 7460 562 6538 1900 25338 26217 57954 27361 31286 4525 1583 32131 2559 5882 2452 5564 16672 5440 1165 1003 7234 1104 1126  19,196 300 9,356 33,853 32,200 0 121,417 1,658 41,840 31,763 3,653 54,761 8,391 98,071 15,480 24,088 195,391 7,145 1,895 163,970 449 18,787 827 16,020 3,600 82,292 85,753 259,054 154,639 106,165 10,270 2,888 109,639 5,159 14,099 4,899 13,183 49,632 12,829 1,994 1,664 18,102 1,869 1,914  1 1 1 1 1 3 3 3 1 1 1 1 1 1 1 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 1 2 2 2 1  134  Table C.1 ctd River Distance (km N of Hope)  269 268 266 266 264 263 263 262 262 261 261 260 260 259 259 259 259 258 257 257 257 256 253 253 252 251 249 249 244 244 243 241 241 240 239 239 238 237 237 236 235 235 235 234  Position (UTM zone 10) Easting Northing  553264 553269 554784 554729 556115 556882 556607 556855 556991 556718 556681 557422 557131 557540 557499 557488 557776 557882 558441 558370 558060 558724 559414 559379 559617 560497 561631 562201 566143 565830 566361 566538 567142 566772 567471 567176 567970 568863 568864 569392 570658 570668 570339 571629  5684886 5683681 5682149 5682285 5681471 5680721 5681060 5679347 5679910 5678704 5678968 5677670 5678233 5677449 5677465 5677486 5676706 5676298 5675218 5675360 5675511 5675282 5672533 5672776 5671924 5670494 5670302 5669917 5667835 5667945 5667203 5665027 5665338 5664650 5663941 5663494 5662657 5662064 5662362 5661595 5661391 5661306 5661496 5661057  Mine Type  Volume Estimation Method  Area (m2 )  Volume (m3 )  Mapping Confidence  S S S S GS S S S S S S S S GS GS GS S S S S S S GS GS RB S H H S S S S S S GS S S S S S GS GS S GS  S S S S GS/H S S S S S S S S GS/H GS/H GS/H S S S S S S GS/H GS/H NV Srv. GS/H Srv. S S S S S S GS/H Srv. Srv. S S S GS/H Srv. S GS/H  166 1696 341 2108 94630 3684 27352 7320 4596 151 11396 43266 3386 184 169 528 14360 1292 43323 871 3325 2050 1536 2632 37999 7391 81572 143265 2109 1131 13168 31557 14945 57370 5048 82718 50514 1393 2362 3038 17440 22828 592 1704  189 3,139 452 4,082 657,332 8,012 90,257 18,362 10,465 169 31,344 157,060 7,235 453 410 1,548 41,442 2,259 157,310 1,403 7,078 3,946 5,379 10,080 0 168,640 552,813 974,460 4,084 1,924 37,323 107,277 43,490 220,847 21,543 222,285 70,176 2,475 4,683 6,347 91,458 57,089 880 6,071  3 1 1 1 1 3 1 3 1 2 1 1 1 2 2 2 1 1 1 1 1 1 2 2 3 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1  135  Table C.1 ctd River Distance (km N of Hope)  234 234 234 234 234 233 233 232 230 229 228 226 225 223 223 222 221 221 221 221 221 221 221 220 220 217 215 211 210 209 208 208 207 207 207 207 207 207 207 207 207 205 204 204  Position (UTM zone 10) Easting Northing  571588 571566 571184 570994 571262 571801 571984 572147 572965 573543 573584 574073 574523 575609 575185 576394 575997 576312 576376 576500 576535 576469 576544 576314 576259 577565 579267 579174 579024 579192 579445 579438 580001 579987 579902 579955 579863 579825 580162 579790 579885 579689 579181 579406  5661088 5661122 5661252 5661593 5660919 5660859 5660626 5660202 5658111 5657827 5656369 5654574 5653867 5651808 5652664 5651204 5650961 5650255 5650363 5650627 5650883 5651030 5651075 5649475 5650164 5647296 5645700 5642535 5640894 5640774 5639819 5639956 5638245 5638288 5638327 5638495 5638250 5638439 5638509 5638614 5638487 5636502 5635469 5635982  Mine Type  Volume Estimation Method  Area (m2 )  Volume (m3 )  Mapping Confidence  GS GS GS S S GS GS S S S S S S H H S GS S S S S S S S S S S S S S S S GS GS GS S S S S S GS S S S  GS/H GS/H Srv. S Srv. GS/H Srv. S S S S S S GS/H Srv. S Srv. S S S S S S S S S S S S S S S GS/H GS/H GS/H S Srv. Srv. Srv. Srv. Srv. S S S  757 408 2344 39336 25293 8067 97294 75127 23609 55830 11913 36021 3950 47007 39577 12293 183792 1813 1446 10490 3958 3639 4502 35746 937 240 4247 771 29652 26830 3662 29400 1582 2266 1951 3090 8297 1605 13819 1014 4632 5910 1590 5714  2,357 1,146 4,032 139,993 71,579 37,217 550,032 305,887 75,558 213,706 33,070 125,869 8,715 290,678 390,643 34,348 382,802 3,402 2,589 28,359 8,737 7,893 10,207 124,709 1,533 296 9,513 1,211 99,504 88,181 7,954 98,484 5,567 8,465 7,109 6,478 11,631 1,001 73,241 783 4,156 14,180 2,903 13,614  1 1 1 1 1 1 1 1 1 1 1 1 1 2 1 1 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1  136  Table C.1 ctd River Distance (km N of Hope)  203 203 200 199 197 197 197 197 194 191 190 188 188 188 188 188 187 187 187 187 186 186 186 186 185 185 185 184 181 181 181 181 181 180 176 176 176 176 176 176 175 175 175 175  Position (UTM zone 10) Easting Northing  Mine Type  Volume Estimation Method  Area (m2 )  Volume (m3 )  Mapping Confidence  578927 579141 579324 578833 578924 579346 579106 579005 581187 581254 581647 580107 579906 580218 580275 580287 579745 579777 579342 579067 578696 578594 578431 578407 578344 578221 578420 577459 575144 574743 575500 574991 575279 575223 576677 576741 576557 576794 576475 576768 576255 576398 576342 576150  S S S S GS H S S H S S H H H H S H H S S H RB? S S S S S S S S S S/RB S/RB GS H H H S S S H H S S  S S S S GS/H Srv. Srv. Srv. GS/H S DE GS/H GS/H GS/H Srv. DE GS/H GS/H S S GS/H DE S S S S S S S S S DE DE GS/H GS/H GS/H DE S S S GS/H GS/H S S  2196 4407 2212 972 690 45119 2761 5224 81674 15487 52046 5446 8839 2659 21468 15360 1405 3506 8868 4198 5116 17635 2563 7878 18786 10646 4775 100620 18419 7577 8490 2408 105836 82418 5648 5196 28797 7712 8684 19788 28565 4360 34637 67937  4,288 9,948 4,326 1,602 2,115 242,289 8,124 9,549 553,619 45,403 5,205 23,536 41,402 10,201 106,696 15,360 4,848 14,083 23,151 9,381 21,882 17,635 5,169 20,066 57,331 28,870 10,960 435,354 55,981 19,144 21,964 2,408 105,836 559,505 24,558 22,281 155,504 19,557 22,572 61,045 162,603 18,159 120,051 270,885  1 1 1 1 1 2 1 1 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 1 1 1 1 1 2 1 1 1 1  5634802 5635231 5632063 5630776 5629658 5629633 5629473 5629504 5627094 5624904 5623492 5622165 5622207 5622233 5622472 5622420 5621928 5622097 5621955 5622040 5622867 5623169 5622937 5623074 5623297 5623300 5623546 5624169 5622720 5623206 5623219 5622951 5622981 5622207 5618278 5618336 5618118 5617852 5617962 5618529 5617719 5617944 5616966 5617253  137  Table C.1 ctd River Distance (km N of Hope)  175 175 174 174 174 173 173 173 172 172 171 171 171 171 171 170 170 155 155 154 154 154 153 151 150 149 148 148 148 148 148 147 147 146 146 146 145 143 143 142 139 139 139 139  Position (UTM zone 10) Easting Northing  576667 576363 575155 576095 575577 576211 575925 575505 575871 576030 576685 576542 576688 576582 576640 576842 576853 584416 584482 584582 584561 584554 585654 586480 587386 588024 588230 588225 588062 588090 588183 589251 588778 589346 589730 589406 590110 590584 590753 590914 592633 592564 592483 592343  5617445 5617853 5616337 5616361 5616320 5615767 5615954 5616032 5614584 5615292 5613943 5614286 5613794 5614048 5614051 5613418 5613521 5601434 5602210 5600588 5600773 5601009 5599747 5598666 5598060 5597733 5597163 5597216 5597636 5597510 5597602 5596420 5596398 5596149 5595335 5596221 5595528 5593183 5593321 5592617 5589820 5589871 5589884 5589857  Mine Type  Volume Estimation Method  Area (m2 )  Volume (m3 )  Mapping Confidence  S S GS H S GS H S S S GS GS S S S GS S GS S H H H S S S S GS GS GS S S S S S S S S S S S GS GS GS S  S S NV GS/H S GS/H Srv. S S DE GS/H GS/H S S S GS/H S GS/H S GS/H GS/H Srv. DE S S S GS/H GS/H GS/H S S S DE S Srv. Srv. DE S S S Srv. Srv. Srv. S  17843 14436 33646 8005 21522 24830 11780 11230 33189 231379 9356 9601 12129 12941 12240 8119 7411 35620 4631 13483 8212 23341 21458 122900 34951 974 576 520 1024 3606 41912 56432 144485 62497 51074 15527 32514 51451 60854 239076 1144 5066 4564 5267  53,873 41,707 0 36,883 67,565 138,088 75,129 30,793 114,015 717,275 44,240 45,594 33,796 36,547 34,170 37,497 18,639 210,340 10,562 67,747 37,998 207,204 21,458 554,343 121,367 1,606 1,714 1,521 3,352 7,807 151,142 216,492 144,485 244,905 390,144 8,073 32,514 193,626 237,149 1,238,430 8,664 24,791 11,092 12,338  1 1 1 1 1 1 1 2 3 1 1 1 1 3 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 1 2 1 1 1 1 1 1 1 1 1 1  138  Table C.1 ctd River Distance (km N of Hope)  138 138 136 135 135 135 135 135 133 132 132 132 132 132 131 130 128 125 125 124 124 123 123 121 120 119 118 118 117 117 117 117 116 115 110 109 108 107 107 107 107 106 105 102  Position (UTM zone 10) Easting Northing  Mine Type  Volume Estimation Method  Area (m2 )  Volume (m3 )  Mapping Confidence  592702 592555 593016 593259 593104 593603 593512 593154 593859 593919 593927 594041 594331 594040 594136 594350 593902 595801 595584 595852 595304 595648 595768 596465 596843 597532 598334 597901 598588 598540 598274 598339 599193 599084 600972 601433 601261 601211 601193 600768 601261 600898 601370 601430  GS S S GS GS S S S H H H S S S S S S GS S/GS GS/S S GS H GS S S S S H H H H S S S S GS GS GS H H GS GS H  GS/H S S GS/H Srv. S S S GS/H GS/H GS/H S S Srv. Srv. S S GS/H Srv. GS/H S GS/H NV GS/H S S S S GS/H GS/H GS/H Srv. Srv. S Srv. S GS/H GS/H GS/H GS/H GS/H GS/H GS/H GS/H  133 46120 21524 7526 22527 1143 4130 32012 12046 4046 8812 14673 18245 114183 37704 37985 34193 5014 82290 39912 41700 5295 23536 3296 40337 17898 32638 104076 6606 4909 4915 15262 564927 18712 37781 79837 51240 14523 10660 44323 37851 41607 1805 3093  310 169,660 67,573 34,322 121,836 1,949 9,197 109,149 59,403 16,643 41,255 42,536 55,343 773,019 70,239 134,206 118,194 21,374 554,039 240,183 150,219 22,777 0 13,104 144,308 54,074 111,732 453,481 29,481 20,853 20,882 208,856 1,651,433 57,059 112,622 329,202 321,428 73,879 51,511 271,416 225,782 252,120 6,493 12,168  1 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 3 1 1 1 1 1 2 2 2 1 1 1 1 1 2 2 2 3 2 3 1 2  5589793 5589481 5587017 5586224 5586274 5586568 5586594 5586639 5584139 5583986 5584047 5583739 5583903 5583232 5582369 5581305 5579737 5577700 5577459 5576180 5576684 5575659 5574892 5573935 5573070 5571596 5571033 5571092 5570062 5570156 5570481 5570304 5568765 5568249 5564194 5562938 5562236 5561680 5561832 5561339 5561088 5560694 5559750 5557030  139  Table C.1 ctd River Distance (km N of Hope)  99 99 99 99 98 98 97 97 95 95 91 91 87 85 85 85 84 83 81 76 75 73 73 71 71 70 69 69 69 69 68 66 66 66 66 66 66 65 62 61 59 55 55 48  Position (UTM zone 10) Easting Northing  602551 602341 602153 602120 602908 602879 602524 602594 602589 602854 603654 603553 605113 605626 605882 605322 606154 606363 606744 609290 610065 610577 610732 611222 611556 611208 611577 611502 611410 611721 611490 611835 611689 611349 611323 611320 611286 611821 612688 612635 611207 611791 611514 613306  5554054 5554122 5554315 5554341 5553086 5553407 5552068 5552677 5550563 5550720 5546987 5547238 5542862 5540815 5540935 5541333 5540488 5539766 5537079 5534188 5532707 5531085 5530904 5529416 5529638 5527921 5526748 5526899 5527036 5527093 5525736 5524270 5524365 5524381 5524452 5524495 5524633 5523302 5520381 5519855 5519132 5515143 5515528 5509307  Mine Type  Volume Estimation Method  Area (m2 )  Volume (m3 )  Mapping Confidence  S S S S S S S S RB S S S S S S S S S S S GS H S H H H H H H H H H H H H H H S S S S S S S  S S S S S S S S NV S S S S S S S S S S S GS/H GS/H S GS/H GS/H GS/H GS/H GS/H GS/H GS/H Srv. GS/H GS/H GS/H GS/H GS/H GS/H S S S S S S S  11168 10847 620 2210 2039 4043 10794 7576 7820 14831 6802 13603 19337 13119 18476 21196 20820 18111 9995 35969 4123 7167 1766 4344 8196 126962 6400 2440 27015 6035 207004 30249 6727 2675 2198 934 2763 121914 17360 5742 14278 8366 6458 6840  30,588 29,529 931 4,321 3,921 8,964 29,355 19,141 0 43,090 16,804 38,818 59,369 37,156 56,190 66,331 64,912 54,852 26,751 125,650 17,013 32,421 3,296 18,081 37,912 926,073 28,412 9,228 152,360 26,531 747,284 173,836 30,112 10,272 8,170 3,011 10,668 548,975 52,117 13,695 41,156 21,578 15,783 16,918  1 2 1 2 3 1 2 3 1 3 1 3 1 2 2 2 3 2 1 3 2 2 3 3 2 1 1 1 1 2 1 1 1 2 2 2 2 1 3 3 3 3 3 2  140  Table C.1 ctd River Distance (km N of Hope)  47 47 46 44 44 43 43 42 41 40 40 40 40 40 38 37 36 33 30 28 24 23 22 19 18 16 15 14 12 12 12 12 12 12 12 11 11 9 9 9 8 6  Position (UTM zone 10) Easting Northing  Mine Type  Volume Estimation Method  Area (m2 )  Volume (m3 )  Mapping Confidence  613829 613662 614054 614702 614936 614633 614630 614547 614299 615105 615053 615014 614974 614944 615192 615307 616064 614716 615285 615752 613156 612702 613531 614580 614670 614457 614572 614569 614551 614561 614566 614537 614536 614511 614486 614516 614593 613530 613776 613807 613677 614348  S S S S S S S S S GS GS GS GS GS S S S S S S S S S/RB S S S S GS GS GS GS GS GS GS S GS GS GS S S S S  S S S S S S S S S GS/H GS/H GS/H GS/H GS/H S S S S S S S S DE Srv. S S S GS/H GS/H GS/H GS/H GS/H GS/H GS/H S GS/H GS/H GS/H S Srv. S S  20107 18221 6418 24762 20938 14915 21786 1188 17664 1286 600 482 571 585 13414 3958 64696 14708 13589 20382 209924 12545 76629 168811 6453 49250 20944 6613 907 542 339 479 114 2978 1555 26850 29784 21184 3154 35128 60881 95070  62,236 55,255 15,665 80,037 65,357 43,385 68,568 2,042 53,221 4,372 1,797 1,392 1,696 1,745 38,167 8,737 255,352 42,658 38,770 63,266 1,058,403 35,201 229,887 1,269,459 15,769 183,665 65,379 29,517 2,910 1,596 923 1,382 259 11,642 2,826 151,276 170,724 114,742 6,641 180,250 237,276 406,515  2 2 2 2 3 1 3 2 1 1 1 2 2 2 2 2 1 1 3 2 3 3 1 1 2 1 1 1 1 1 1 1 1 1 0 2 1 1 1 1 3 3  5508348 5508829 5507447 5506131 5506379 5505257 5505509 5504601 5503333 5501856 5501932 5501991 5502049 5502101 5500436 5499776 5498926 5496627 5494071 5491919 5490568 5489911 5488848 5486043 5485192 5482814 5481935 5481571 5478917 5478974 5479011 5479086 5479236 5479511 5479602 5478407 5478675 5477187 5476539 5476923 5475670 5474212  141  Map 2.1 Historical Minesites Along Fraser River From Hope to Quesnel Forks  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items