Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Multi-year hydrologic response of experimental waste-rock piles in a cold climate : active-zone development,… Fretz, Nathan Mackenzie 2013

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

Item Metadata

Download

Media
24-ubc_2013_spring_fretz_nathan.pdf [ 14.75MB ]
Metadata
JSON: 24-1.0073576.json
JSON-LD: 24-1.0073576-ld.json
RDF/XML (Pretty): 24-1.0073576-rdf.xml
RDF/JSON: 24-1.0073576-rdf.json
Turtle: 24-1.0073576-turtle.txt
N-Triples: 24-1.0073576-rdf-ntriples.txt
Original Record: 24-1.0073576-source.json
Full Text
24-1.0073576-fulltext.txt
Citation
24-1.0073576.ris

Full Text

MULTI-YEAR HYDROLOGIC RESPONSE OF EXPERIMENTAL WASTE-ROCK PILES IN A COLD CLIMATE: ACTIVE-ZONE DEVELOPMENT, NET INFILTRATION, AND FLUID FLOW  by Nathan Mackenzie Fretz  B.Sc., The University of British Columbia, 2010  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Geological Sciences)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  February 2013  © Nathan Mackenzie Fretz, 2013  Abstract A large-scale field study was constructed at the Diavik Diamond Mines, NT, Canada, to examine the hydrologic, geochemical, microbiologic, gas-transport, and heat-transport mechanisms that influence drainage water quality from waste-rock. This thesis focuses on the hydrology component of the project, with the aim of better understanding fluid flow through mine waste-rock at multiple scales in cold climates. The hydrologic response of two largescale, experimental waste-rock piles (test piles) and a series of smaller-scale lysimeters (active-zone lysimeters) were monitored. Both the test piles and the active-zone lysimeters were constructed on top of water collection systems in order to monitor outflows, and were instrumented with tensiometers, moisture content sensors, and thermistors. Active-zone development, net infiltration, matrix wet-up, wetting front movement, and outflow at the test piles and active-zone lysimeters are examined. Above-freezing air temperatures result in an active-zone developing inwards into the test piles during the summer months, allowing net infiltration at the thawed surface and fluid flow through thawed regions to occur. Over a five year period the average percent net infiltration into the active-zone lysimeters and crowns of the test piles was estimated to be 43%, 57%, 15%, 41%, and 58% of natural rainfall in 2007, 2008, 2009, 2010, and 2011, respectively. An indirect evapotranspiration computation suggests that net infiltration from rainfall events <5 mm/d is restricted to periods early (May) and late (September to mid-October) in the rainseason, while net infiltration in June, July, and August is restricted to rainfall events >5 mm/d. TDR sensors are used to monitor initial wet-up of the matrix fraction and wetting front velocities. Approximate spatial uniformity of wetting front velocities for all but the most intense rainfall events, and positive correlation between outflow rate and  ii  solute/dissolved metal concentration from effluent samples, suggest that the dominant mechanism of fluid flow at the test piles and active-zone lysimeters is porewater displacement in the matrix from propagating pressure waves, as opposed to preferential mechanisms. Over a 5 year period outflow below the batters of the test piles was significantly greater than outflow below the crowns, even after the matrix material below the crowns completely wet-up.  iii  Table of Contents  Abstract .................................................................................................................................... ii Table of Contents ................................................................................................................... iv List of Tables ........................................................................................................................... x List of Figures ........................................................................................................................ xii Acknowledgements ........................................................................................................... xxxii Chapter 1: Introduction ........................................................................................................ 1 1.1  Background and significance .................................................................................... 1  1.2  Diavik waste-rock project ......................................................................................... 3  1.3  Scope and organization of thesis............................................................................... 3  1.4  Literature review – hydrogeology of waste-rock ...................................................... 5  1.4.1  Stockpile construction ........................................................................................... 5  1.4.2  Fluid flow through waste-rock .............................................................................. 6  1.4.2.1  Matrix Flow .................................................................................................. 6  1.4.2.2  Preferential Flow ........................................................................................... 7  1.4.2.3  Flow mechanism interaction ......................................................................... 8  1.4.3  Hydraulic conductivity.......................................................................................... 9  1.4.4  Permafrost and waste-rock .................................................................................. 10  1.5  Figures..................................................................................................................... 12  Chapter 2: Experimental set-up and previous work ........................................................ 13 2.1  Site description........................................................................................................ 13  2.2  Experiment construction and design ....................................................................... 14  iv  2.2.1  15 m test piles ..................................................................................................... 14  2.2.1.1  Drainage collection system ......................................................................... 16  2.2.2  Active-zone lysimeters (AZLs)........................................................................... 18  2.2.3  Field-permeameter .............................................................................................. 19  2.3  Instrumentation ....................................................................................................... 20  2.3.1  Meteorological station ........................................................................................ 21  2.3.2  Rain gauge tipping buckets ................................................................................. 21  2.3.3  Tensiometers ....................................................................................................... 21  2.3.4  Soil moisture sensors .......................................................................................... 22  2.3.4.1  ECH2O sensors............................................................................................ 22  2.3.4.2  Time domain reflectrometry (TDR) sensors ............................................... 23  2.3.4.3  Comments on moisture content and matric potential sensors ..................... 24  2.3.5  Lysimeter locations ............................................................................................. 25  2.3.6  Flow cells ............................................................................................................ 26  2.3.7  Tipping buckets ................................................................................................... 26  2.3.8  Dataloggers and telemetry .................................................................................. 27  2.4  Key values from previous work .............................................................................. 27  2.4.1  Bulk waste-rock properties ................................................................................. 28  2.4.1.1  Grain-size distribution ................................................................................ 28  2.4.1.2  Hydraulic properties.................................................................................... 28  2.4.2  Matrix-fraction (<5 mm) properties .................................................................... 30  2.4.2.1  Grain-size distribution ................................................................................ 30  2.4.2.2  Hydraulic properties.................................................................................... 30  v  2.4.3  Field-scale experiments (test piles and active zone lysimeters) ......................... 31  2.4.3.1  Field capacity and initial moisture content ................................................. 31  2.4.3.2  Infiltration capacity ..................................................................................... 32  2.5  Concluding statement.............................................................................................. 32  2.6  Figures..................................................................................................................... 34  2.7  Tables ...................................................................................................................... 50  Chapter 3: Infiltration into waste-rock in a northern climate ........................................ 53 3.1  Introduction ............................................................................................................. 53  3.1.1 3.2  ET studies in waste-rock ..................................................................................... 55 Methods................................................................................................................... 57  3.2.1  Study site and meteorological instrumentation ................................................... 57  3.2.2  Active zone lysimeter (AZLs) water balance ..................................................... 57  3.2.3  Moisture content and matric potential sensors.................................................... 59  3.2.4  FAO-56 Penman-Monteith method .................................................................... 59  3.2.4.1  Estimating reference evapotranspiration..................................................... 60  3.2.4.2  Estimating actual evaporation ..................................................................... 61  3.2.4.3  Estimating net infiltration ........................................................................... 63  3.3  Results and discussion ............................................................................................ 64  3.3.1  Climate ................................................................................................................ 64  3.3.2  Net infiltration ..................................................................................................... 66  3.3.2.1  Active-zone lysimeter water balance .......................................................... 66  3.3.2.2  FAO-56 Penman-Monteith method ............................................................ 67  3.3.2.3  Factors influencing net infiltration.............................................................. 68  vi  3.3.2.4  Application of net infiltration estimates to the crowns of the uncovered test  piles  70  3.3.3  Uncertainty, limitations, and sensitivity ............................................................. 72  3.4  Conclusions ............................................................................................................. 74  3.5  Figures..................................................................................................................... 76  3.6  Tables ...................................................................................................................... 89  Chapter 4: Wet-up and fluid flow in waste-rock undergoing annual freeze-thaw ........ 93 4.1  Introduction ............................................................................................................. 93  4.2  Methods................................................................................................................... 94  4.2.1  Test piles ............................................................................................................. 94  4.2.2  Active-zone lysimeters........................................................................................ 97  4.3  Results and Discussion ........................................................................................... 98  4.3.1  Active-zone formation ........................................................................................ 98  4.3.1.1  Active-zone lysimeters................................................................................ 98  4.3.1.2  Test piles ..................................................................................................... 99  4.3.2  Initial matrix wet-up ......................................................................................... 101  4.3.2.1  Active-zone lysimeters.............................................................................. 101  4.3.2.2  Type III test pile ........................................................................................ 102  4.3.2.3  Type I test pile .......................................................................................... 104  4.3.3  Fluid flow .......................................................................................................... 105  4.3.4  Type I vs. Type III test pile matrix wet-up ....................................................... 108  4.3.5  Outflow ............................................................................................................. 110  4.3.5.1  Active-zone lysimeters.............................................................................. 110  vii  4.3.5.2  Test piles ................................................................................................... 112  4.3.5.3  Centre versus batter outflow ..................................................................... 116  4.4  Conclusions ........................................................................................................... 118  4.5  Figures................................................................................................................... 121  4.6  Tables .................................................................................................................... 142  Chapter 5: Conclusions and recommendations .............................................................. 145 5.1  Conclusions ........................................................................................................... 145  5.2  Recommendations ................................................................................................. 149  References ............................................................................................................................ 151 Appendices ........................................................................................................................... 155 Appendix A Hydrology dataset ........................................................................................ 155 A.1  Rainfall .............................................................................................................. 155  A.2  Air temperature ................................................................................................. 162  A.3  Type I test pile .................................................................................................. 167  A.4  Type III test pile ................................................................................................ 184  A.5  Covered test pile ............................................................................................... 224  A.6  Active-zone lysimeters (AZLs)......................................................................... 243  A.7  FAO-56 Penman-Monteith evaporation and net infiltration analysis ............... 268  Appendix B Tipping buckets ............................................................................................ 274 B.1  Design ............................................................................................................... 274  B.2  Calibration......................................................................................................... 279  Appendix C Field-permeameter ........................................................................................ 281 C.1  Draindown test .................................................................................................. 281  viii  C.2  Constant head tests ............................................................................................ 282  Appendix D MATLAB analysis scripts ............................................................................ 287 D.1  Time domain reflectrometry analysis example ................................................. 288  D.2  Tensiometer example ........................................................................................ 295  D.3  Tipping bucket example .................................................................................... 297  ix  List of Tables  Table ‎2-1  Test pile and active zone lysimeter (AZL) dimensions ...................................... 50  Table ‎2-2  Surviving time domain reflectrometry (TDR) sensor labels and datalogger IDs 50  Table ‎2-3  Test pile basal drain and basal collection lysimeter labels and datalogger IDs . 51  Table ‎2-4  Bulk waste-rock properties – Field-permeameter experiments .......................... 51  Table ‎2-5  Matrix material hydraulic properties .................................................................. 52  Table ‎2-6  Miscellaneous field-scale experiment parameters .............................................. 52  Table ‎3-1  Key parameters used in calculating Ke ............................................................... 89  Table ‎3-2  Natural rainfall measured at the crowns of the test piles .................................... 89  Table ‎3-3  Applied rainfall at the crown of the Type III test pile (area of influence =  20mx30m) ............................................................................................................................. 89 Table ‎3-4  Applied rainfall at the Type III active-zone lysimeters (AZLs) ......................... 90  Table ‎3-5  Total annual rainfall at the test piles and active-zone lysimeters (AZLs) .......... 90  Table ‎3-6  Rainfall (natural) statistics .................................................................................. 90  Table ‎3-7  Net infiltration (NI) calculated based on active-zone lysimeter water balances 91  Table ‎3-8  Percent net infiltration (PNI) calculated based on active-zone lysimeter water  balances  ............................................................................................................................. 91  Table ‎3-9  Comparison of AZL and FAO-PM net infiltration (NI) and percent net  infiltration (PNI) ..................................................................................................................... 91 Table ‎3-10  First-order estimates of net infiltration (NI) and percent net infiltration (PNI)  into the crowns of the uncovered test piles ............................................................................. 92 Table ‎3-11  FAO-PM method sensitivity analysis (using 2008 natural rainfall) ............... 92  x  Table ‎4-1  Dates for the onset of thawed and frozen conditions at the active zone lysimeters ........................................................................................................................... 142  Table ‎4-2  Active zone lysimeter outflow volumes ........................................................... 142  Table ‎4-3  Basal drain outflow volumes at the uncovered test piles .................................. 143  Table ‎4-4  Interpolated basal drain discharge at the uncovered test piles .......................... 144  Table ‎4-5  Centre versus batter outflow at the Type III test pile........................................ 144  Table ‎A-1  Natural rainfall measured at the crowns of the test piles .............................. 161  Table ‎A-2  Total annual rainfall at the test piles and active-zone lysimeters (AZLs) .... 161  Table ‎A-3  Applied rainfall at the crown of the Type III test pile (area of influence =  20mx30m)  ....................................................................................................................... 161  Table ‎A-4  Applied rainfall at the Type III active-zone lysimeters (AZLs) ................... 161  Table ‎A-5  Basal drain outflow volumes at the Type I test pile...................................... 183  Table ‎A-6  Basal collection lysimeter outflow volumes at the Type I test pile .............. 183  Table ‎A-7  Basal drain outflow volumes at the Type III test pile ................................... 222  Table ‎A-8  Basal collection lysimeter outflow volumes at the Type III test pile ........... 222  Table ‎A-9  Basal drain outflow volumes at the Covered test pile................................... 242  Table ‎A-10  Basal drain outflow volumes at the Covered test pile................................... 267  Table ‎A-11  FAO-PM method results ............................................................................... 273  Table ‎B-1 Tipping bucket calibration equations for 2008 through 2011 ........................... 280 Table ‎C-1 Constant head test results .................................................................................. 286  xi  List of Figures  Figure ‎1-1  Location of the Diavik Diamond Mine (64o29’‎N,‎110o18’‎W) ..................... 12  ‎Figure ‎1-2  Conceptual waste-rock stockpile. Construction features include lifts, traffic  surfaces, and gravitational segregation. Adapted from Fala et al. (2005) .............................. 12 Figure ‎2-1  Location of the Diavik Diamond Mine (64o29’‎N,‎110o18’‎W) ..................... 34  Figure ‎2-2  Diavik Diamond Mine on East Island ............................................................. 34  Figure ‎2-3  Aerial photograph of the Test Piles Research Area ........................................ 35  Figure ‎2-4  Schematic of the Test Piles Research Area. Red denotes instrument huts at the  base of the piles. Green denotes instrument huts at the crowns of the piles. Re-created and adapted from Smith (2009) ..................................................................................................... 36 Figure ‎2-5  Pre-construction design of the Type I (a), Type III (b) and Covered (c) test  piles. Solid black lines denote the basal extent of each pile. Dotted black lines denote the crowns of each pile. Dashed red lines denote the basal drain line and the arrows denote flow direction. Rectangular cells are locations of basal collection lysimeters. Re-created from Smith (2009) ......................................................................................................................... 37 Figure ‎2-6  Post-construction drawings of the Type I test pile and drainage scheme. (a):  white represents the crown, grey represents the batters, and the square patterns represent the basal collection lysimeters. (b): contour interval is 0.2 m. Adapted from Neuner (2009) ..... 38 Figure ‎2-7  Post-construction drawings of the Type III test pile and drainage scheme. (a):  white represents the crown, grey represents the batters, and the square patterns represent the basal collection lysimeters (BCLs). (b): contour interval is 0.2 m. Adapted from Neuner (2009)  ......................................................................................................................... 39  xii  Figure ‎2-8  During- and post-construction pictures of the active zone lysimeters (AZLs).  (a): Empty water drums and drainlines. From foreground to background the water drums represent the Type III West, Type III East, Type I West, and Type I East AZLs. (b): AZLs being filled with run-of-mine material. (c): surface of the completed AZLs. (a) and (b) sourced from Neuner (2009) ................................................................................................... 40 Figure ‎2-9  Design of the 32 m3 field-permeameter. Middle image shows the slotted drain  at the base of the field-permeameter. Image source: Momeyer (in progress) ........................ 41 Figure ‎2-10  Field-permeameter - experiment set-up .......................................................... 41  Figure ‎2-11  Rain gauge tipping bucket at the crown of the Type I test pile ...................... 42  Figure ‎2-12  Custom-built time domain reflectrometry (TDR) sensors. Image source:  Neuner (2009) ......................................................................................................................... 43 Figure ‎2-13  North-South cross-section of the Type I test pile, 2 m west of the centre line.  TDR sensor locations denoted by red dots. Adapted from Smith (2009) ............................... 43 Figure ‎2-14 line.  (a): West-East cross-section of the Type III test pile, 2 m south of the centre  (b): West-East cross-section of the Type III test pile, 2 m north of the centre line.  TDR sensor locations denoted by red dots. Adapted from Smith (2009) ............................... 44 Figure ‎2-15 line.  (a): South-North cross-section of the Covered test pile, 2 m east of the centre  (b): South-North cross-section of the Covered test pile, 2 m west of the centre line.  TDR sensor locations denoted by red dots. Adapted from Smith (2009) ............................... 45 Figure ‎2-16  Relative locations of the basal collection lysimeters (BCLs) at the base of the  Type I test pile. Consult figures 2-5 and 2-6 for overall location within the test pile. ........... 46 Figure ‎2-17  Relative locations of the basal collection lysimeters (BCLs) at the base of the  Type III test pile. Consult figures 2-5 and 2-7 for overall location within the test pile. ........ 46  xiii  Figure ‎2-18  Relative locations of the basal collection lysimeters (BCLs) at the base of the  Covered test pile. Consult figure 2-5 for overall location within the test pile. ....................... 46 Figure ‎2-19  Flow-cell set-up at the Covered test pile basal drain outflow (CBxxdrn13) .. 47  Figure ‎2-20  Uninstalled custom-built tipping bucket manufactured at UBC ..................... 47  Figure ‎2-21  Young Model 2202 tipping bucket raingauges at the Type III BCL outflow . 48  Figure ‎2-22  Grain-size distribution of the run-of-mine waste-rock. Source: Neuner (2009) . ......................................................................................................................... 48  Figure ‎2-23  (a): Water retention curves of matrix material. (b): Hydraulic conductivity  curves of matrix material. Source: Neuner et al. (2012) ......................................................... 49 Figure ‎3-1  Type I test pile crown, facing north. (a): Snow covered crown on March 21,  2012.  ......................................................................................................................... 76  Figure ‎3-2  Type III test pile crown, facing west. (a): Snow covered crown on March 21,  2012.  ......................................................................................................................... 77  Figure ‎3-3  Surface texture at the upper surface of the AZLs (a), Type I test pile crown  (b), and Type III test pile crown (c). Columns illustrate similar surface textures of the different locations. The white square in each image is 1 m on each side. .............................. 78 Figure ‎3-4  2010 average monthly incoming solar radiation recorded at the DDMI  meteorological station. Source: DDMI Environment Dept. ................................................... 79 Figure ‎3-5  2010 air temperature at the DDMI meteorological station ............................. 79  Figure ‎3-6  Daily rainfall measured at the crowns of the test piles in 2010 (a) and 2011 (b) ......................................................................................................................... 80  Figure ‎3-7  FAO-PM method output for the Type I active-zone lysimeters in 2008 ........ 81  xiv  Figure ‎3-8  Comparison of yearly active-zone lysimeter (AZL) net infiltration estimates  and FAO-PM calculated net infiltration ................................................................................. 82 Figure ‎3-9  FAO-PM calculated net infiltration and Type I active –zone lysimeter (AZL)  outflow for 2008. AZL outflow at the end of the year is approximately equal to net infiltration.  ......................................................................................................................... 83  Figure ‎3-10  FAO-PM calculated net infiltration and Type III active-zone lysimeter (AZL)  outflow for 2008. AZL outflow at the end of the year is approximately equal to net infiltration.  ......................................................................................................................... 83  Figure ‎3-11  FAO-PM calculated net infiltration and active-zone lysimeter (AZL) outflow  for 2009. AZL outflow at the end of the year is approximately equal to net infiltration. Note: Different primary y-axis scale than Figures 3-9, 3-10, 3-12, and 3-13. ................................. 84 Figure ‎3-12  FAO-PM calculated net infiltration and active-zone lysimeter (AZL) outflow  for 2010. AZL outflow at the end of the year is approximately equal to net infiltration. ...... 84 Figure ‎3-13  FAO-PM calculated net infiltration and active-zone lysimeter (AZL) outflow  for 2011. AZL outflow at the end of the year is approximately equal to net infiltration. ...... 85 Figure ‎3-14  Type I West active-zone lysimeter (AZL) outflow and near surface volumetric  moisture content response at the uncovered test pile crowns for 2008................................... 86 Figure ‎3-15  Active-zone lysimeter (AZL) outflows and near surface volumetric moisture  content response at the uncovered test pile crowns for 2009.................................................. 86 Figure ‎3-16  Active-zone lysimeter (AZL) outflows and near surface volumetric moisture  content response at the uncovered test pile crowns for 2010.................................................. 87 Figure ‎3-17  Active-zone lysimeter (AZL) outflows and near surface volumetric moisture  content response at the uncovered test pile crowns for 2011.................................................. 87  xv  Figure ‎3-18  FAO-PM calculated net infiltration and flux past tensiometer profiles at the  crown of the Type III test pile for 2007. Truncation of tensiometer data represents removal of instrumentation. ...................................................................................................................... 88 Figure ‎3-19  FAO-PM method sensitivity analysis using the depth of the evaporation layer  (Ze) and the maximum depth of water that can be evaporated from the evaporation layer without restriction (REW). The best-fit net infiltration corresponds to the best-fit Ze and REW values for all years. ....................................................................................................... 88 Figure ‎4-1  North-South cross-section of the Type I test pile, 2 m west of the centre line.  TDR sensor locations denoted by red dots. Adapted from Smith (2009) ............................. 121 Figure ‎4-2 line.  (a): West-East cross-section of the Type III test pile, 2 m south of the centre  (b): West-East cross-section of the Type III test pile, 2 m north of the centre line.  TDR sensor locations denoted by red dots. Adapted from Smith (2009) ............................. 122 Figure ‎4-3  Active zone lysimeter and air temperatures. The 0.6 m ECH2O sensor was not  recoverable after significant wire damage was sustained at the end of September, 2009. ... 123 Figure ‎4-4  Comparison of 31S2 TDR apparent volumentric moisture content and 31S5  thermistor temperature. ......................................................................................................... 124 Figure ‎4-5  Comparison of 12W2 TDR 9 m apparent volumentric moisture content and  12W5 thermistor 9 m temperature. ....................................................................................... 125 Figure ‎4-6  Apparent volumentric moisture content at the 31S2 TDR string (Type III test  pile) for 2009 ....................................................................................................................... 126 Figure ‎4-7  Volumetric moisture content at the 31S2 TDR string in the Type III test pile ... ....................................................................................................................... 127  xvi  Figure ‎4-8  Volumetric moisture content at the 33N2 TDR string in the Type III test pile .. ....................................................................................................................... 128  Figure ‎4-9  1 m depth TDR sensors in the Type III test pile ........................................... 129  Figure ‎4-10  7 m depth TDR sensors in the Type III test pile ........................................... 130  Figure ‎4-11  9 m depth TDR sensors in the Type III test pile ........................................... 131  Figure ‎4-12  1 m depth TDR sensors in the Type III test pile from 2006 through 2007 ... 132  Figure ‎4-13  7 m depth TDR sensors in the Type III test pile during 2007 ....................... 132  Figure ‎4-14  Volumetric moisture content at the 11W2and 12W2 TDR strings in the Type I  test pile  ....................................................................................................................... 133  Figure ‎4-15  2007 through 2011 cumulative annual outflow (normalized by surface  collection area) at the active-zone lysimeters ....................................................................... 134 Figure ‎4-16  2011 daily outflow at the Type I East (a) and Type I West (b) active-zone  lysimeters  ....................................................................................................................... 135  Figure ‎4-17  Type I test pile main basal drain outflow...................................................... 136  Figure ‎4-18  Type III test pile main basal drain outflow ................................................... 137  Figure ‎4-19  Outflow from the 3BNBlys2W/2E Type III basal collection lysimeters and  apparent volumetric moisture contents along the 31S2 TDR string ..................................... 138 Figure ‎4-20  Outflow from the 3BNClys4E Type III basal collection lysimeter and apparent  volumetric moisture contents along the 31S2 TDR string .................................................... 139 Figure ‎4-21  2011 Type III basal drain outflow ................................................................ 140  Figure ‎4-22  2011 3BNClys2W/2E outflow and 31S2 TDR apparent volumetric moisture  contents  ....................................................................................................................... 140  Figure ‎4-23  Type I and Type III basal drain outflow ....................................................... 141  xvii  Figure ‎A-1  2007 through 2011 natural rainfall at the Test Piles Research Area (all test  piles and active-zone lysimeters) .......................................................................................... 156 Figure ‎A-2  2007 natural rainfall at the Type I test pile, Type III test pile batters, Covered  test pile, and Type I active-zone lysimeters .......................................................................... 157 Figure ‎A-3  2007 natural and applied rainfall at the crown of the Type III test pile ........ 157  Figure ‎A-4  2007 natural and applied rainfall at the Type III active-zone lysimeters ..... 158  Figure ‎A-5  2008 natural rainfall at the Type I test pile, Type III test pile, Covered test  pile, and Type I active-zone lysimeters ................................................................................ 158 Figure ‎A-6  2008 natural and applied rainfall at the Type III active-zone lysimeters ..... 159  Figure ‎A-7  2009 natural rainfall for the Test Piles Research Area ................................. 159  Figure ‎A-8  2010 natural rainfall for the Test Piles Research Area ................................. 160  Figure ‎A-9  2011 natural rainfall for the Test Piles Research Area ................................. 160  Figure ‎A-10  2007 through 2011 air temperature at the Test Piles Research Area ............ 163  Figure ‎A-11  2007 air temperature at the Test Piles Research Area .................................. 164  Figure ‎A-12  2008 air temperature at the Test Piles Research Area .................................. 164  Figure ‎A-13  2009 air temperature at the Test Piles Research Area .................................. 165  Figure ‎A-14  2010 air temperature at the Test Piles Research Area .................................. 165  Figure ‎A-15  2011 air temperature at the Test Piles Research Area .................................. 166  Figure ‎A-16  2007 through 2011 apparent volumetric moisture content at the Type I test  pile  ....................................................................................................................... 168  Figure ‎A-17  2007 apparent volumetric moisture content at the Type I test pile ............... 169  Figure ‎A-18  2008 apparent volumetric moisture content at the Type I test pile ............... 169  Figure ‎A-19  2009 apparent volumetric moisture content at the Type I test pile ............... 170  xviii  Figure ‎A-20  2010 apparent volumetric moisture content at the Type I test pile ............... 170  Figure ‎A-21  2011 apparent volumetric moisture content at the Type I test pile ............... 171  Figure ‎A-22  2007 through 2011 basal drain outflow at the Type I test pile ..................... 172  Figure ‎A-23  2007 basal drain outflow at the Type I test pile ............................................ 173  Figure ‎A-24  2008 basal drain outflow at the Type I test pile ............................................ 173  Figure ‎A-25  2009 basal drain outflow at the Type I test pile ............................................ 174  Figure ‎A-26  2010 basal drain outflow at the Type I test pile ............................................ 174  Figure ‎A-27  2011 basal drain outflow at the Type I test pile ............................................ 175  Figure ‎A-28  2007 through 2011 1BWBlys Cluster basal collection lysimeter outflow at the  Type I test pile ...................................................................................................................... 176 Figure ‎A-29  2009 1BWBlys Cluster basal collection lysimeter outflow at the Type I test  pile  ....................................................................................................................... 177  Figure ‎A-30  2010 1BWBlys Cluster basal collection lysimeter outflow at the Type I test  pile  ....................................................................................................................... 177  Figure ‎A-31  2007 through 2011 1BWClys Cluster basal collection lysimeter outflow at the  Type I test pile ...................................................................................................................... 178 Figure ‎A-32  2010 1BWClys Cluster basal collection lysimeter outflow at the Type I test  pile  ....................................................................................................................... 179  Figure ‎A-33  2007 through 2011 1BEClys Cluster basal collection lysimeter outflow at the  Type I test pile. Note the scale. ............................................................................................. 180 Figure ‎A-34  2008 1BEClys Cluster basal collection lysimeter outflow at the Type I test  pile  ....................................................................................................................... 181  xix  Figure ‎A-35  2009 1BEClys Cluster basal collection lysimeter outflow at the Type I test  pile  ....................................................................................................................... 181  Figure ‎A-36  2010 1BEClys Cluster basal collection lysimeter outflow at the Type I test  pile  ....................................................................................................................... 182  Figure ‎A-37  2011 1BEClys Cluster basal collection lysimeter outflow at the Type I test  pile  ....................................................................................................................... 182  Figure ‎A-38  2006 through 2011 apparent volumetric moisture content along TDR-string  31N2 at the Type III test pile ................................................................................................ 185 Figure ‎A-39  2007 apparent volumetric moisture content along TDR-string 31N2 at the  Type III test pile .................................................................................................................... 186 Figure ‎A-40  2008 apparent volumetric moisture content along TDR-string 31N2 at the  Type III test pile .................................................................................................................... 186 Figure ‎A-41  2009 apparent volumetric moisture content along TDR-string 31N2 at the  Type III test pile .................................................................................................................... 187 Figure ‎A-42  2010 apparent volumetric moisture content along TDR-string 31N2 at the  Type III test pile .................................................................................................................... 187 Figure ‎A-43  2009 apparent volumetric moisture content along TDR-string 31N2 at the  Type III test pile .................................................................................................................... 188 Figure ‎A-44  2006 through 2011 apparent volumetric moisture content along TDR-string  31S2 at the Type III test pile ................................................................................................. 189 Figure ‎A-45  2007 apparent volumetric moisture content along TDR-string 31S2 at the  Type III test pile .................................................................................................................... 190  xx  Figure ‎A-46  2008 apparent volumetric moisture content along TDR-string 31S2 at the  Type III test pile .................................................................................................................... 190 Figure ‎A-47  2009 apparent volumetric moisture content along TDR-string 31S2 at the  Type III test pile .................................................................................................................... 191 Figure ‎A-48  2010 apparent volumetric moisture content along TDR-string 31S2 at the  Type III test pile .................................................................................................................... 191 Figure ‎A-49  2011 apparent volumetric moisture content along TDR-string 31S2 at the  Type III test pile .................................................................................................................... 192 Figure ‎A-50  2006 through 2011 apparent volumetric moisture content along TDR-string  33N2 at the Type III test pile ................................................................................................ 193 Figure ‎A-51  2007 apparent volumetric moisture content along TDR-string 33N2 at the  Type III test pile .................................................................................................................... 194 Figure ‎A-52  2008 apparent volumetric moisture content along TDR-string 33N2 at the  Type III test pile .................................................................................................................... 194 Figure ‎A-53  2009 apparent volumetric moisture content along TDR-string 33N2 at the  Type III test pile .................................................................................................................... 195 Figure ‎A-54  2010 apparent volumetric moisture content along TDR-string 33N2 at the  Type III test pile .................................................................................................................... 195 Figure ‎A-55  2011 apparent volumetric moisture content along TDR-string 33N2 at the  Type III test pile .................................................................................................................... 196 Figure ‎A-56  2006 through 2011 apparent volumetric moisture content along TDR-string  33S2 at the Type III test pile ................................................................................................. 197  xxi  Figure ‎A-57  2007 apparent volumetric moisture content along TDR-string 33S2 at the  Type III test pile .................................................................................................................... 198 Figure ‎A-58  2008 apparent volumetric moisture content along TDR-string 33S2 at the  Type III test pile .................................................................................................................... 198 Figure ‎A-59  2009 apparent volumetric moisture content along TDR-string 33S2 at the  Type III test pile .................................................................................................................... 199 Figure ‎A-60  2010 apparent volumetric moisture content along TDR-string 33S2 at the  Type III test pile .................................................................................................................... 199 Figure ‎A-61  2011 apparent volumetric moisture content along TDR-string 33S2 at the  Type III test pile .................................................................................................................... 200 Figure ‎A-62  2007 through 2011 basal drain outflow at the Type III test pile (North and  South drains combined) ........................................................................................................ 201 Figure ‎A-63  2007 basal drain outflow at the Type III test pile (North and South drains  combined)  ....................................................................................................................... 202  Figure ‎A-64  2008 basal drain outflow at the Type III test pile (North and South drains  combined)  ....................................................................................................................... 202  Figure ‎A-65  2009 basal drain outflow at the Type III test pile (North and South drains  combined)  ....................................................................................................................... 203  Figure ‎A-66  2010 basal drain outflow at the Type III test pile (North and South drains  combined)  ....................................................................................................................... 203  Figure ‎A-67  2011 basal drain outflow at the Type III test pile (North and South drains  combined)  ....................................................................................................................... 204  Figure ‎A-68 ........................................................................................................................... 205  xxii  Figure ‎A-69  2007 through 2011 3BNBlys2W/2E basal collection lysimeter outflow at the  Type III test pile .................................................................................................................... 206 Figure ‎A-70  2009 3BNBlys2W/2E basal collection lysimeter outflow at the Type III test  pile  ....................................................................................................................... 207  Figure ‎A-71  2010 3BNBlys2W/2E basal collection lysimeter outflow at the Type III test  pile  ....................................................................................................................... 207  Figure ‎A-72  2007 through 2011 3BNBlys4W basal collection lysimeter outflow at the  Type III test pile .................................................................................................................... 208 Figure ‎A-73  2009 3BNBlys4W basal collection lysimeter outflow at the Type III test pile .. ....................................................................................................................... 209  Figure ‎A-74  2010 3BNBlys4W basal collection lysimeter outflow at the Type III test pile .. ....................................................................................................................... 209  Figure ‎A-75  2011 3BNBlys4W basal collection lysimeter outflow at the Type III test pile .. ....................................................................................................................... 210  Figure ‎A-76  2011 3BNClys2W/2E basal collection lysimeter outflow at the Type III test  pile  ....................................................................................................................... 211  Figure ‎A-77  2011 3BNClys2W/2E basal collection lysimeter outflow at the Type III test  pile  ....................................................................................................................... 212  Figure ‎A-78  2007 through 2011 3BNClys4W basal collection lysimeter outflow at the  Type III test pile .................................................................................................................... 213 Figure ‎A-79  2008 3BNClys4W basal collection lysimeter outflow at the Type III test pile .. ....................................................................................................................... 214  xxiii  Figure ‎A-80  2007 through 2011 3BNClys4E basal collection lysimeter outflow at the Type  III test pile  ....................................................................................................................... 215  Figure ‎A-81  2008 3BNClys4E basal collection lysimeter outflow at the Type III test pile ... ....................................................................................................................... 216  Figure ‎A-82  2009 3BNClys4E basal collection lysimeter outflow at the Type III test pile ... ....................................................................................................................... 216  Figure ‎A-83  2010 3BNClys4E basal collection lysimeter outflow at the Type III test pile ... ....................................................................................................................... 217  Figure ‎A-84  2011 3BNClys4E basal collection lysimeter outflow at the Type III test pile ... ....................................................................................................................... 217  Figure ‎A-85  2007 through 2011 3BSClys4E basal collection lysimeter outflow at the Type  III test pile  ....................................................................................................................... 218  Figure ‎A-86  2008 3BSClys4E basal collection lysimeter outflow at the Type III test pile .... ....................................................................................................................... 219  Figure ‎A-87  2009 3BSClys4E basal collection lysimeter outflow at the Type III test pile .... ....................................................................................................................... 219  Figure ‎A-88  2007 through 2011 3BSClys4E basal collection lysimeter outflow at the Type  III test pile  ....................................................................................................................... 220  Figure ‎A-89  2010 hydraulic head calculated from tensiometers at the Type III test pile  crown  ....................................................................................................................... 221  Figure ‎A-90  2011 hydraulic head calculated from tensiometers at the Type III test pile  crown  ....................................................................................................................... 221  xxiv  Figure ‎A-91  2008 through 2011 apparent volumetric moisture content along TDR-string  C2E2 at the Covered test pile................................................................................................ 225 Figure ‎A-92  2008 apparent volumetric moisture content along TDR-string C2E2 at the  Covered test pile ................................................................................................................... 226 Figure ‎A-93  2009 apparent volumetric moisture content along TDR-string C2E2 at the  Covered test pile ................................................................................................................... 226 Figure ‎A-94  2010 apparent volumetric moisture content along TDR-string C2E2 at the  Covered test pile ................................................................................................................... 227 Figure ‎A-95  2011 apparent volumetric moisture content along TDR-string C2E2 at the  Covered test pile ................................................................................................................... 227 Figure ‎A-96  2008 through 2011 apparent volumetric moisture content along TDR-string  C2W2 at the Covered test pile .............................................................................................. 228 Figure ‎A-97  2008 apparent volumetric moisture content along TDR-string C2W2 at the  Covered test pile ................................................................................................................... 229 Figure ‎A-98  2009 apparent volumetric moisture content along TDR-string C2W2 at the  Covered test pile ................................................................................................................... 229 Figure ‎A-99  2010 apparent volumetric moisture content along TDR-string C2W2 at the  Covered test pile ................................................................................................................... 230 Figure ‎A-100  2011 apparent volumetric moisture content along TDR-string C2W2 at the  Covered test pile ................................................................................................................... 230 Figure ‎A-101  2008 through 2011 apparent volumetric moisture content along TDR-string  C3E2 at the Covered test pile................................................................................................ 231  xxv  Figure ‎A-102  2008 apparent volumetric moisture content along TDR-string C3E2 at the  Covered test pile ................................................................................................................... 232 Figure ‎A-103  2009 apparent volumetric moisture content along TDR-string C3E2 at the  Covered test pile ................................................................................................................... 232 Figure ‎A-104  2010 apparent volumetric moisture content along TDR-string C3E2 at the  Covered test pile ................................................................................................................... 233 Figure ‎A-105  2011 apparent volumetric moisture content along TDR string C3E2 at the  Covered test pile ................................................................................................................... 233 Figure ‎A-106  2007 through 2011 basal drain outflow at the Covered test pile .............. 234  Figure ‎A-107  2007/2008 season basal drain outflow at the Covered test pile ................ 235  Figure ‎A-108  2008/2009 season basal drain outflow at the Covered test pile ................ 235  Figure ‎A-109  2009/2010 season basal drain outflow at the Covered test pile ................ 236  Figure ‎A-110  2009/2011/2012 season basal drain outflow at the Covered test pile.  Outflow still occurring after truncation in December, 2011. ................................................ 237 Figure ‎A-111  2010 hydraulic head calculated from tensiometers at the Covered test pile  crown  ................................................................................................................... 238  Figure ‎A-112  2011 hydraulic head calculated from tensiometers at the Covered test pile  crown  ................................................................................................................... 238  Figure ‎A-113  2009 temperature and volumetric moisture content (VMC) within the Type  III material and till at the Covered test pile at STA 0+94 ..................................................... 239 Figure ‎A-114  2010 temperature and volumetric moisture content (VMC) within the Type  III material and till at the Covered test pile at STA 0+94 ..................................................... 239  xxvi  Figure ‎A-115  2009 temperature and volumetric moisture content (VMC) at the top of the  till layer at the Covered test pile, on the batters downslope of the crown ............................ 240 Figure ‎A-116  2010 temperature and volumetric moisture content (VMC) at the top of the  till layer at the Covered test pile, on the batters downslope of the crown ............................ 240 Figure ‎A-117  2009 temperature and volumetric moisture content (VMC) at the bottom of  the till layer at the Covered test pile, on the batters downslope of the crown ...................... 241 Figure ‎A-118  2010 temperature and volumetric moisture content (VMC) at the bottom of  the till layer at the Covered test pile, on the batters downslope of the crown ...................... 241 Figure ‎A-119  2007 through 2011 temperature and volumetric moisture content (VMC)  from ECH2O sensors at the active zone-lysimeters .............................................................. 244 Figure ‎A-120  2007 temperature and volumetric moisture content (VMC) from ECH2O  sensors at the active zone-lysimeters .................................................................................... 245 Figure ‎A-121  2008 temperature and volumetric moisture content (VMC) from ECH2O  sensors at the active zone-lysimeters .................................................................................... 245 Figure ‎A-122  2009 temperature and volumetric moisture content (VMC) from ECH2O  sensors at the active zone-lysimeters .................................................................................... 246 Figure ‎A-123  2010 temperature and volumetric moisture content (VMC) from ECH2O  sensors at the active zone-lysimeters .................................................................................... 246 Figure ‎A-124  2011 temperature and volumetric moisture content (VMC) from ECH2O  sensors at the active zone-lysimeters .................................................................................... 247 Figure ‎A-125  2010 hydraulic head calculated from tensiometers at the active zone-  lysimeters  ................................................................................................................... 248  xxvii  Figure ‎A-126  2011 hydraulic head calculated from tensiometers at the active zone-  lysimeters  ................................................................................................................... 248  Figure ‎A-127  2007 through 2011 outflow at the Type I East active-zone lysimeter ...... 249  Figure ‎A-128  2008 outflow at the Type I East active-zone lysimeter............................. 250  Figure ‎A-129  2009 outflow at the Type I East active-zone lysimeter............................. 250  Figure ‎A-130  2010 outflow at the Type I East active-zone lysimeter............................. 251  Figure ‎A-131  2011 outflow at the Type I East active-zone lysimeter............................. 251  Figure ‎A-132  2007 through 2011 outflow at the Type I West active-zone lysimeter ..... 252  Figure ‎A-133  2008 outflow at the Type I West active-zone lysimeter ........................... 253  Figure ‎A-134  2009 outflow at the Type I West active-zone lysimeter ........................... 253  Figure ‎A-135  2010 outflow at the Type I West active-zone lysimeter ........................... 254  Figure ‎A-136  2011 outflow at the Type I West active-zone lysimeter ........................... 254  Figure ‎A-137  2007 through 2011 outflow at the Type III East active-zone lysimeter ... 255  Figure ‎A-138  2007 outflow at the Type III East active-zone lysimeter .......................... 256  Figure ‎A-139  2008 outflow at the Type III East active-zone lysimeter .......................... 256  Figure ‎A-140  2009 outflow at the Type III East active-zone lysimeter .......................... 257  Figure ‎A-141  2010 outflow at the Type III East active-zone lysimeter .......................... 257  Figure ‎A-142  2011 outflow at the Type III East active-zone lysimeter .......................... 258  Figure ‎A-143  2007 through 2011 outflow at the Type III West active-zone lysimeter .. 259  Figure ‎A-144  2007 outflow at the Type III West active-zone lysimeter......................... 260  Figure ‎A-145  2008 outflow at the Type III West active-zone lysimeter......................... 260  Figure ‎A-146  2009 outflow at the Type III West active-zone lysimeter......................... 261  Figure ‎A-147  2010 outflow at the Type III West active-zone lysimeter......................... 261  xxviii  Figure ‎A-148  2011 outflow at the Type III West active-zone lysimeter......................... 262  Figure ‎A-149  2007 through 2011 cumulative annual outflow (normalized by surface  collection area) at the active-zone lysimeters ....................................................................... 263 Figure ‎A-150  2007 cumulative annual outflow (normalized by surface collection area) at  the active-zone lysimeters ..................................................................................................... 264 Figure ‎A-151  2008 cumulative annual outflow (normalized by surface collection area) at  the active-zone lysimeters ..................................................................................................... 264 Figure ‎A-152  2009 cumulative annual outflow (normalized by surface collection area) at  the active-zone lysimeters ..................................................................................................... 265 Figure ‎A-153  2010 cumulative annual outflow (normalized by surface collection area) at  the active-zone lysimeters ..................................................................................................... 265 Figure ‎A-154  2011 cumulative annual outflow (normalized by surface collection area) at  the active-zone lysimeters ..................................................................................................... 266 Figure ‎A-155  2007 FAO-PM method calculation, using rainfall at the crown of the Type I  test pile  ................................................................................................................... 269  Figure ‎A-156  2008 FAO-PM method calculation, using rainfall at the crown of the Type  III test pile  ................................................................................................................... 269  Figure ‎A-157  2008 FAO-PM method calculation, using rainfall at the Type I active-zone  lysimeters  ................................................................................................................... 270  Figure ‎A-158  2008 FAO-PM method calculation, using rainfall at the Type III active-  zone lysimeters ................................................................................................................... 270 Figure ‎A-159  2009 FAO-PM method calculation, using rainfall representative of all areas  in the Test Piles Research Area ............................................................................................ 271  xxix  Figure ‎A-160  2010 FAO-PM method calculation, using rainfall representative of all areas  in the Test Piles Research Area ............................................................................................ 271 Figure ‎A-161  2011 FAO-PM method calculation, using rainfall representative of all areas  in the Test Piles Research Area ............................................................................................ 272 Figure ‎B-1  AutoCad drawing of the tipping bucket box design. Original design by Neuner  (2009), updated by Momeyer (in progress). ......................................................................... 275 Figure ‎B-2  AutoCad drawing of the tipping bucket design. Original design by Neuner  (2009), updated by Momeyer (in progress). ......................................................................... 276 Figure ‎B-3  Support triangles reinforcing the connections between the side-walls and base  of the tipping bucket box. Source: Modified from Momeyer (in progress). ......................... 277 Figure ‎B-4  Bumpers to cushion the bucket during tips and level adjusters to facilitate easy  levelling of the bucket. Source: Modified from Momeyer (in progress). ............................. 277 Figure ‎B-5  Non-threaded drainage system at the base of the tipping bucket box. Source:  Modified from Momeyer (in progress). ................................................................................ 278 Figure ‎B-6  Calibration curve for the 1Bxxdrn13 (Type I test pile basal drain) tipping  bucket. Calibration was conducted on July 19th, 2011. ........................................................ 280 Figure ‎C-1  Field-permeameter draindown test showing flow rate measured at the base of  the field-permeameter and watertable level within the field-permeameter .......................... 284 Figure ‎C-2  Field-permeameter draindown test showing volume of outflow as a percent of  the total initial volume of water taken to saturate the field-permeameter and watertable level within the field-permeameter ................................................................................................ 285  xxx  List of Abbreviations  α  A van Genuchten (1980) shape parameter for water retention curves  θres  Volumetric moisture content at residual saturation  θsat  Volumetric moisture content at saturation  AMD  Acid mine drainage  AZL  Active-zone lysimeter  BCL  Basal collection lysimeter  DDMI  Diavik Diamond Mines Inc.  FAO  Food and Agricultural Organization of the United Nations  HDPE  High-density poly ethylene  Ksat  Saturated hydraulic conductivity  η  A van Genuchten (1980) shape parameter for water retention curves  NAG  Non-acid generating  PAG  Potentially acid generating  PVC  Polyvinyl chloride  REV  Representative elemental volume  SWSS  Soil water solution sampler  TDR  Time domain reflectrometry  UTP  Uncovered test piles (Type I and III)  VMC  Volumetric moisture content  VMCa  Apparent volumetric moisture content  WRC  Wetting retention curve  xxxi  Acknowledgements  To…  …Mom‎and‎Dad. Without your encouragement, love, and support, I would never have applied‎to‎this‎M.Sc.‎and‎been‎supervised‎by…  …Leslie‎Smith.‎Thank you for your fantastic supervision, open door, and accessibility regardless of travel. If not for your involvement in the Diavik Waste-Rock Project, I would never‎have‎met…‎  ...David Blowes, Richard Amos, and Dave Sego. Thank you for your leadership and advice throughout my degree; your individual knowledge was invaluable. and... …Lianna Smith, Matt Lindsay, Brenda Bailey, Stacey Hannam, Jeff Bain, Thomas Eckert, Nam Pham, and Sean Sinclair. Thank you for all your help and for always making me feel welcome at Waterloo. and… …my‎fellow grad students of rm. 59, 60, 227, and 229. Thank you for your friendship.  To Matthew Neuner and Steven Momeyer: Thank you for laying the groundwork for the hydrology of the test piles, for your support in getting me up to speed with the project, and for answering all my questions... even late into my degree.  xxxii  The research presented in this thesis is part of a larger, collaborative project: The Diavik Waste-Rock Project. Collaborators include the University of Alberta, the University of British Columbia, the University of Waterloo, and the Diavik Diamond Mines (DDMI). Additional funding for the project has been provided by the Natural Sciences and Engineering Research Council (CRD program), the Canadian Foundation for Innovation, the Mine and Environment Neutral Drainage Program (MEND), and the International Network for Acid Prevention (INAP). I would like to thank the people at Diavik who helped me onsite, particularly Dave Mohler, Steve Pinter, Dan Guignon, and Richard LeBreton.  xxxiii  Chapter 1: Introduction  1.1  Background and significance Environmental issues associated with waste-rock at mine sites primarily involve the  release of acidity, sulfate, and metals to the surrounding environment. This process is commonly referred to as Acid Mine Drainage (AMD) and results from the weathering of mine-wastes that contain sulfide minerals, the most common of which are pyrite and pyrrhotite. Oxidation of sulfide minerals occurs when they are exposed to atmospheric oxygen and water, a process that is microbiologically mediated. The oxidation of sulfide minerals generates acidity and releases sulfate and metals into solution, which may be subsequently transported via fluid flow through the waste-rock and released to the environment. AMD is a coupled physical and biogeochemical process, which means that the prediction of solute loading to the environment requires not only an understanding of the biogeochemical processes that control sulfide mineral weathering, but also an understanding of the hydrologic processes and fluid flow mechanisms that control unsaturated flow through waste-rock (e.g. Lefebvre, 2001; Smith and Beckie, 2003; Stockwell et al., 2006). Much attention has been given to understanding the geochemical and mineralogical composition of mine-waste, in order to predict whether or not a stockpile will release environmentally damaging effluent. It is correspondingly important to place focus towards understanding the processes of fluid flow within mine-waste,‎since‎it‎is‎“central‎to‎the‎design‎of‎an‎effective‎ monitoring strategy to characterize metal release and in the interpretation of field data collected‎within,‎beneath,‎or‎adjacent‎to‎an‎actively‎weathering‎pile”‎(Smith‎and‎Beckie,‎ 2003).  1  Understanding and characterizing fluid flow through waste-rock is challenging due to the extremely heterogeneous nature of the material. Standard field methods for determining hydraulic parameters, such as hydraulic conductivity, are not possible in unsaturated stockpiles. Instead, measurements of hydraulic properties are primarily carried out at the laboratory scale on small volumes of material consisting of the fine grained fraction of the waste-rock. For example, Tempe cell and lab-permeameter tests may be conducted to develop water retention curves and determine hydraulic properties of the fine grained fraction of the waste-rock. Using these lab-scale measurements to describe field-scale behaviour can be problematic. Fluid flow through soil-like waste-rock that is dominated by fine grained material may be reasonably described by the hydraulic properties measured at the lab-scale; however, waste-rock with a broader grain-size distribution that includes cobbles and boulders may exhibit fluid flow mechanisms not observable at the lab-scale, such as preferential flow through larger macropores. The difference in fluid flow mechanisms between the lab- and field-scale‎is‎known‎as‎the‎‘scale‎problem’‎and‎is‎a‎main‎concern‎when‎using‎lab-scale measurements to predict fluid flow at the field-scale. Previous and concurrent studies have examined the hydrology and fluid flow mechanisms of unsaturated, heterogeneous waste-rock at both the lab-scale and field-scale (e.g. at the Antamina Mine, Peru and the Cluff Lake Mine, Saskatchewan). However, the Diavik Waste-Rock Project (the subject of this thesis) is unique in that a large-scale field experiment is being conducted for waste-rock deposited onto permafrost, where fluid flow is disrupted by annual freeze-thaw cycles and is restricted to an active-zone that develops during the summer months. The scarcity of literature on the hydrogeologic behaviour of waste-rock in cold climates, and the need to better understand the mechanisms of fluid flow  2  through waste-rock deposited onto permafrost, are driving forces behind this research at the Diavik Diamond Mines, NT, Canada.  1.2  Diavik waste-rock project The Diavik Diamond Mine is located approximately 300 km northeast of  Yellowknife, NT, Canada (Figure 1-1). The research presented in this thesis is part of a larger, collaborative project: The Diavik Waste-Rock Project, which is examining the hydrologic, geochemical, microbiologic, gas-transport, and heat-transport mechanisms that influence drainage water quality from waste-rock. The project aims to describe the physical, geochemical, and microbiological processes affecting the weathering of waste-rock in large stockpiles in cold climates and to explore scaling from laboratory measurements to fieldscale behaviour.  1.3  Scope and organization of thesis The hydrology component of the Diavik Waste-Rock Project focuses on the examination  of hydraulic parameters, fluid flow mechanisms, and water balances of waste-rock at multiple scales. The aim is to better understand fluid flow through waste-rock deposited onto permafrost and to determine the scale-up relationship between lab-scale measurements and field-scale fluid flow response. Work by Neuner (2009) focused on the early phases of the study, including waste-rock characterization and the initial water accumulation phase of the test‎piles.‎This‎thesis‎builds‎on‎Neuner’s‎research,‎as‎the test piles approached and entered a state of hydrogeological dynamic equilibrium. The specific research goals follow:  3  1. Quantify net infiltration at the test piles and determine the periods and conditions most important for net infiltration of water into waste-rock in a northern climate. 2. Assess the applicability of an indirect computational method for estimating evaporation and net infiltration at the test piles (FAO Penman-Monteith recommended method). 3. Investigate time-scales for stockpile wet-up in a northern climate. 4. Investigate wetting front movement following various intensity rainfall events. 5. Investigate the evolution of test pile outflow over annual and multi-year time-scales. 6. Quantify the proportion of test pile outflow that originated below the batters (batter outflow) versus below the crowns (centre outflow) of the test piles.  This thesis consists of five chapters. Chapter 1 is introductory and includes background information on acid mine drainage, the Diavik Waste-Rock Project, and the hydrogeology of waste-rock. Chapter 2 describes the study site, the experimental set-up and design, and presents key findings from previous work on the project essential to the analysis contained within this thesis. The next two chapters compose the body of the thesis and are written in article format with an introduction, methods (additional to Chapter 2), results and discussion, and conclusions section. Chapter 3 focuses on net infiltration, while Chapter 4 describes the freeze-thaw, initial matrix wet-up, water movement, and outflow at the uncovered test piles. Chapter 5 compiles the key findings and conclusions and includes recommendations for future work. The Covered test pile is not addressed in the body of the thesis and is left to future work. A number of appendices document the full hydrology dataset from 2007 through 2011 (Appendix A; includes Covered test pile data), tipping bucket design and calibration  4  (Appendix B), field-permeameter experiment (Appendix C), and MATLAB data analysis scripts (Appendix D).  1.4  1.4.1  Literature review – hydrogeology of waste-rock  Stockpile construction Mine waste-rock is a by-product of open-pit and underground mining, whereby non-  ore rock that is blasted and excavated to access ore bodies and rock that is below the economic grade for processing are stockpiled (Blowes et al., 2003). Mining method, pile construction method, and waste-rock lithology can result in a particle-size distribution of the waste-rock material ranging from clays to boulders (Smith and Beckie, 2003; Blowes et al., 2003). Stockpiles range from tens to hundreds of metres in height and are commonly constructed in a series of lifts using end-dumping and push-dumping techniques. These construction techniques result in the gravitational segregation of the waste-rock based on particle size; the waste-rock is dumped down a series of slopes (or tip faces) and settles at the angle of repose with the average grain-size increasing towards the base of the stockpile. As a result, bulk-permeability may also increase towards the base of the stockpile. Traffic surfaces can occur internal to and on the upper-most surface of waste-rock piles due to heavy equipment movement during lift construction, leading to thin zones of decreased permeability. The particle-size distribution of waste-rock, the segregation of waste-rock during pile construction, and the surfaces and internal structures formed during pile construction result in significant heterogeneity of waste-rock piles. In addition, waste-rock piles are emplaced in an  5  unsaturated state; which, combined with heterogeneity, makes it challenging to characterize the hydrogeological properties of the stockpiles. Figure 1-2 is a conceptual cross-section of a stockpile, showing internal structure and grain-size segregation.  1.4.2  Fluid flow through waste-rock Smith and Beckie (2003) define two factors as having principal control on the  hydrologic properties of waste-rock piles: 1) the grain-size distribution of the waste-rock, and 2) the proportion and spatial arrangement of matrix-supported and matrix-free zones within the stockpile. The latter is extremely difficult to characterize at any scale and is a principle reason why fluid flow through heterogeneous waste-rock piles is such a challenge to study. The proportion and spatial arrangement of matrix-supported and matrix-free zones are governed by pile construction, which results in: 1) zones of hydraulically connected matrix material (finer grain-sizes) that form due to matrix material settling in the pore spaces between coarser grains (i.e. cobbles and boulders), and 2) large void spaces where pore space between coarse grains are matrix-free and can form isolated or continuous pathways. Both matrix and matrix-free pathways can transmit water, resulting in waste-rock piles with multimodal hydraulic properties. Flow processes through these two permeable pathways are termed‎‘flow‎mechanisms’‎and‎consist‎of‎matrix‎flow‎and‎preferential‎flow.  1.4.2.1  Matrix Flow Matrix flow is governed by capillary and gravity forces acting on the fluid; fine-  grained matrix material can have significant capillarity, giving it the capacity to transport and store water under tension. Matrix flow is slow and diffuse, compared to preferential  6  mechanisms, and occurs in regions of the stockpile dominated by soil-like, granular structure. Several studies have shown cases where matrix flow is the dominant mechanism in wasterock stockpiles (e.g. Wagner et al., 2006; Neuner, 2009; Bay, 2009; Neuner et al., 2012). Fluid flow via capillary forces in unsaturated porous media can be described by the onedimensional Richards equation (Equation 1),  [ ( )(  )]  (1)  where θ‎is‎moisture content, z is depth, K(ψ) is hydraulic conductivity as a function of pressure head (ψ). Yazdani et al. (2000), in developing a method for constructing soil water characteristic curves of mine waste-rock, found that grains greater than 5 mm in diameter exhibit little capillarity. This provides an upper bound for material in which capillary driven matrix flow occurs.  1.4.2.2  Preferential Flow Preferential flow is a non-specific term. It can refer to macropore flow through  connected, matrix-free voids or film flow over the surfaces of large grains. Macropore pathways act to channelize fluid flow, by-passing the matrix and transmitting water very rapidly. Nichol et al. (2005) observed macropore fluid velocities up to several metres per day in a 5 m high waste-rock pile experiment. Flow through macropore pathways is not governed by capillary forces and, consequently, is not described by the Richards equation.  7  Other mechanisms of fluid flow in waste-rock include non-vertical flow, which can occur along dipping stratigraphic units and along fine grained lenses that act as capillary barriers.  1.4.2.3  Flow mechanism interaction The activation of macropore flow is controlled by infiltration rates at the surface of  the waste-rock pile and ponded zones interior to the pile. At low flux rates, fluid flow occurs through the matrix. As matrix moisture content and flux rate increase, ponding can occur and result in free water entering the macropore pathways, which then have higher conductivity than the matrix and can transmit water more rapidly. For example, Nichol et al. (2005) observed macropore flow at flux rates higher than the saturated hydraulic conductivity of the matrix. Multiple studies of waste-rock and macroporous soils indicate that preferential flow is likely a common mechanism of fluid flow in heterogeneous material under the conditions required for initiating preferential flow (e.g. Flury et al., 1994, Smith and Beckie, 2003; Nichol et al., 2005). The proportional contribution of matrix and preferential flow mechanisms and the interactions between them control solute mobilization, solute transport within the piles, and subsequent drainage to the environment. This is because: 1) fluid flow mechanisms and internal pile structure influence the residence time of water in contact with the waste-rock material, and 2) there is a reactive surface area effect with grain size, where the fine grained matrix fraction has the largest reactive surface area. Fluid flowing through the matrix is in contact with higher reactivity material for a longer amount of time than fluid flowing through preferential pathways. Fluid flowing through preferential pathways may  8  have lower solute concentrations due to shorter contact times with lower reactivity material (e.g. Smith and Beckie 2003, Wagner et al. 2006). An understanding of the relationship between fluid flow mechanisms and mineral weathering, as well as knowledge of the specific fluid flow mechanisms occurring in the stockpile of interest are necessary to understand the temporal variation in solute loading to the environment. This is a complex task; Stockwell et al. (2006) demonstrate the difficulty in characterizing fluid flow pathways within waste-rock and then correlating the hydrogeologic properties of the waste-rock with geochemistry.  1.4.3  Hydraulic conductivity Hydraulic conductivity is often the most important hydraulic property in describing  fluid flow through porous media. Unfortunately, it is also extremely difficult to characterize given the heterogeneity and unsaturated state of waste-rock piles. Standard methods for determining saturated hydraulic conductivity are not a possibility in unsaturated waste-rock stockpiles and obtaining a value for saturated hydraulic conductivity only yields one point on the hydraulic conductivity-saturation relationship, which can be hysteretic and multi-modal for heterogeneous material with multiple porosities and fluid pathways. Lab-scale permeameter tests can be used to determine the saturated hydraulic conductivity of the fine grained fraction of the waste-rock, while tempe cells can be used to develop a soil water characteristic curve (SWCC) of the fine grained fraction of the wasterock. From these lab tests, a hydraulic conductivity-saturation relationship can be estimated using various close-formed analytical equations (e.g. van Genuchten 1980, Mualem 1976, Brooks and Corey 1964). Construction of a SWCC using in situ measurements at the fieldscale is also possible using co-located measurements of moisture content and soil suction.  9  The problem with these methods lies in developing a hydraulic conductivity-saturation relationship that applies only to the fine grained fraction of the waste-rock when stockpiles represent multi-modal systems. Multi-modal models exist for describing fluid flow and transport through both matrix and preferential pathways in the unsaturated zone. The most common forms of multi-modal models are dual-porosity (e.g. van Genuchten and Wierenga, 1976) and dual-permeability (e.g. Gerke and van Genuchten, 1993) models. Both models describe the exchange of water and solutes between the matrix and preferential pathways. They differ in that dual-permeability models assume fluid flow occurs in both the matrix and preferential pathways, while dual-porosity models assume fluid in the matrix is stagnant. In situ gas permeability tests can be used to determine permeability of a stockpile (Amos et al., 2009). Perforated air permeability balls installed within stockpiles pick up contributions from all void spaces in the surrounding region of influence, not just matrix pores, enabling the bulk permeability of the stockpile to be measured.  1.4.4  Permafrost and waste-rock Permafrost conditions provide a unique opportunity for northern mines to control  AMD from mine-waste. Permafrost development in waste-rock piles is promoted by stockpiling mine-waste onto permafrost; however, near-surface thawing in the summer months may result in the development of an active zone, where fluid flow can occur and effluent can be discharged to the environment. Mine closure strategies to solve this problem involve employing engineered cover systems to limit the active-zone to a non-acid generating (NAG) cover material, such that a potentially acid generating (PAG) core remains permanently frozen (Smith et al., 2012).  10  Active-zone development in waste-rock stockpiles differs from the development in surrounding porous media; increased permeability and the presence of large, air-filled voids results in the active-zone extending deeper in stockpiles. Matrix material within the stockpile that has wet-up before permafrost develops will be largely ice filled, while larger macropores/voids may remain air-filled and not accumulate ice.  11  1.5  Figures  Figure ‎1-1  Location of the Diavik Diamond Mine (64 o29’‎N,‎110o18’‎W)  ‎0Figure ‎1-2  Conceptual waste-rock stockpile. Construction features include lifts, traffic surfaces,  and gravitational segregation. Adapted from Fala et al. (2005)  12  Chapter 2: Experimental set-up and previous work  2.1  Site description Figure 2-1 shows the location of Diavik Diamond Mine Inc. (DDMI), approximately  300 km northeast of Yellowknife, NT, Canada (64o29’‎N,‎110o18’‎W).‎The‎mine‎is‎situated‎ in the semi-arid, Arctic Tundra and is north of the line of continuous permafrost. Production began in 2003 and current mining practices include both open-pit and underground mining techniques to exploit three kimberlite-ore pipes located on a 20 km2 island (informally East Island) within Lac de Gras (Figure 2-2). The country rock surrounding the kimberlite pipes consists of Archean granite and pegmatite-granite, containing metasedimentary biotite-schist xenoliths (Smith, 2009). DDMI segregates and stockpiles the country rock (waste-rock) based on sulfur content. The biotite-schist is the only rock-type containing more than trace amounts of sulfide minerals; it contains locally disseminated pyrrhotite and other minor sulfides (Smith, 2009), whose abundance controls the management plan for segregation of waste-rock into Type I, Type II, and Type III material. Type I waste-rock is identified as material with an average sulfur content <0.04 wt. %S and is considered non-acid generating (NAG). Type II material is identified as having 0.04 to 0.08 wt. %S and is considered to have low or no acid generating potential. Type III material is identified as having >0.08 wt. %S and is considered potentially acid generating (PAG). At full buildout, DDMI waste-rock stockpiles are projected to contain 200 Mt of waste-rock and have a maximum height of 80 m.  13  2.2  Experiment construction and design The Diavik Waste-Rock Project consists of experiments conducted at multiple scales  including laboratory, 2 m-field, and 15 m-field scales. Figures 2-3 and 2-4 show the Test Piles Research Area at DDMI. The area contains three 15 m high waste-rock experimental piles (test piles), four 2 m high active-zone lysimeters (AZLs), a field-permeameter, and multiple instrumentation huts. This section summarizes construction and design of the Test Piles Research Area as it pertains to hydrology. For a more complete documentation of construction and design, readers are directed to Smith (2009); however, this section contains important points regarding test pile drainage schemes that are not addressed elsewhere.  2.2.1  15 m test piles Three 15 m high test piles were constructed and instrumented between 2005 and  2007. Two of the test piles are distinguished by their sulfur content, whereas the third test pile is distinguished by its construction configuration. The Type I test pile is constructed of NAG material with an average sulfur content of 0.035 wt. %S, while the Type III test pile is constructed of PAG material with an average sulfur content of 0.053 wt. %S1 . The Type I and‎Type‎III‎test‎piles‎are‎both‎referred‎to‎as‎‘uncovered‎test‎piles’.‎The‎Covered‎test‎pile‎has‎ a Type III core constructed of PAG material with a higher average sulfur content of 0.083 wt. %S. This core was re-sloped and overlain by an engineered dry-cover consisting of 1.5 m of till and a further 3 m of Type I waste-rock. The purpose of the cover is to promote permanent freezing of the PAG core, such that low quality drainage is not released to the surrounding environment. The 1.5 m till layer has lower permeability and is designed as a barrier to  1  This‎is‎below‎DDMI’s‎minimum‎threshold of 0.08 wt. %S for Type III waste-rock  14  restrict oxygen and water ingress into the PAG core. The 3 m Type I layer is designed as a thermal freeze-thaw cover to restrict the active zone depth to the Type I material and maintain the PAG core frozen year-round. The Type I and Type III test piles have dimensions of approximately 50 m by 60 m at the base, with batters sloping at the angle of repose of the material (38o or 1.3H:1V). The basal dimensions of the piles were determined such that the crowns of the piles would be a minimum of 20 m wide to accommodate equipment access. Figure 2-5 shows preconstruction design of the test piles. The post-construction crown dimensions of the uncovered test piles are larger than pre-construction design, and each spans an area of approximately 1350 m2. Post construction diagrams are shown in Figures 2-6 and 2-7. The Covered test pile has dimensions of approximately 80 m by 125 m at the base with slopes recontoured to 18o (3H:1V). Figure 2-5 includes a pre-construction design of the Covered test pile. The test piles were constructed on top of their own drainage collection systems (discussed below) by push-dumping and end-dumping techniques, in a series of tip faces. Hydrology instrumentation (section 2.3) was installed at the base of each test pile, as well as along the four tip faces of each test pile. Additionally, the till and Type I layers of the Covered pile were instrumented. Instrument leads were extended out the crowns of the three test piles. The leads were first buried 0.5 m beneath the surface to allow the haul trucks and dozers to continue constructing the test piles outwards by tip faces. Once the test piles were fully constructed, an excavator was used to expose the buried leads and bring them to the surface, disrupting the traffic surface created by the heavy equipment. As a result, a traffic  15  surface is only partially developed at the crowns of the test piles and is likely not representative of the DDMI stockpile surface.  2.2.1.1  Drainage collection system Each test pile is located on top of its own water collection system. Each consists of an  HDPE liner graded to direct water towards a heat-traced, perforated drainpipe and then to custom-built tipping buckets for flow measurement. The liner is overlain by approximately 0.3 m of protective crush material in order to prevent the liner and PVC drainpipe from being damaged during pile construction. The liner is surrounded by 0.5 m high lined-berms, preventing any water that reaches the liner from bypassing the collection system. Postconstruction Figures 2-6 and 2-7 show that the liners do not cover the full extent of the test piles and exclude a small portion of the batter-toes. The water collection systems also include multiple small-scale basal collection lysimeters (BCLs) located on top of the 0.3 m crush layer within each test pile. There are six 4 m by 4 m BCLs and six 2 m by 2 m BCLs in each test pile. Each BCL is lined with HDPE, heat traced to allow water reaching the liner to remain liquid, and graded to direct water through PVC pipe to tipping buckets. The BCLs are intended for the study of spatial variability in flow paths and aqueous geochemistry within the test piles. Pre-construction design aimed to have four BCLs under the batters of each test pile and 8 BCLs under the crown (Figure 2-5); however, post-construction surveys show that both the Type I and Type III test piles each have only two 2 m by 2 m lysimeters under their batters (Figures 2-6 and 2-7). An issue critical to the analysis of the hydrology data in Chapter 4 concerns the difference in test pile drainage schemes. Drainage scheme refers to the layout of the drainage  16  collection systems at the base of the test piles and how water is directed out of the test piles. The Type I and Type III uncovered test piles have different drainage schemes (Figures 2-5 to 2-7), which must be considered when comparing the outflow response from the base of the piles. The contour map in Figure 2-6 shows that water reporting to the base of the Type I test pile flows towards the southwest corner and is directed through the centre of the pile. The contour map in Figure 2-7 shows that water reporting to the base of the Type III test pile flows from the centre to drainpipes that direct effluent to two discharge points located at the northwest and southwest corners. This difference in how water is routed out of the test piles is important because the piles undergo annual freeze-thaw cycles (Chapter 4). As ambient air temperatures rise above freezing during the summer, the test piles begin to thaw inwards and an active-zone develops. The active zone extends through the batters prior to the more central locations, which become active at a later date. Since water is routed to the edges of the Type III test pile, all water reporting to the liner has a thawed pathway along the HDPE towards the drainpipe and subsequent discharge point. However, water reporting to the liner from the batters of the Type I test pile may, depending on location , only report to the drainpipe once a thawed pathway along the HDPE liner has formed in the centre of the test pile (Figure 2-6). Another difference between the Type I and III test pile drainage schemes involves the crush material used to protect the liner and drainpipe. The Type III test pile utilizes 2 inchclean crush, while the Type I test pile utilizes 2 inch-minus crush. The difference reflects material availability at the time of pile construction. The 2 inch-minus crush (Type I test pile) contains fines with the ability to retain water under tension and, therefore, needs to be sufficiently wet-up prior to drainage occurring. By the end of 2011 the Type I test pile had yet to wet-up all the way from the crown to the basal liner (Chapter 4), meaning that any  17  water reporting to the liner below the batters of the Type I test pile had to wet-up the crush material surrounding the drainpipe prior to drainage from the pile occurring. As a result, the Type I test pile may have some amount of water held in storage within the crush material that would otherwise drain out of the pile if 2-inch clean crush had been used. The water held in storage may act as a thermal sink, whereby the presence of an ice mass in the crush material would increase the energy required to thaw the full extent of the drainage system. This could lead to some sections of the crush material surrounding the drainpipe remaining frozen yearround, thus restricting discharge from the pile. The Covered test pile has a drainage scheme similar to the Type I test pile. Water reaching the basal liner is directed to a drainpipe that runs through the centre of the pile towards the northwest corner.  2.2.2  Active-zone lysimeters (AZLs) Four AZLs were constructed in order to study physical, geochemical, and  microbiological processes within the upper 2 m of the waste-rock. Figure 2-8 shows the AZLs during and after construction. They were constructed by filling HDPE tanks with runof-mine waste rock, with boulders >2 m removed. Two of the AZLs were filled with Type I material and two with Type III material. The Type I AZLs have surface collection areas of 3.6 m2 and zero-tension drains at 1.45 m depth. The Type III AZLs have surface collection areas of 2.1 m2 and zero-tension drains at 1.7 m depth. The HDPE tank for the Type I East AZL was damaged during construction, and repaired using HDPE liner. Water reaching each AZL drain is directed through heat-traced PVC pipe to individual tipping buckets for flow measurement. Hydrology instrumentation (Section 2.3) was installed adjacent to the AZLs in  18  similar run-of-mine material. The upper surfaces of the AZLs are at the same elevation as, and within 100 m of, the crowns of the test piles, meaning that they are all exposed to similar climatic conditions. Minimal heavy equipment traffic at the AZLs likely means their surfaces are not comparable to the DDMI stockpiles, but are comparable to the test piles (Chapter 3).  2.2.3  Field-permeameter Two field-permeameters were constructed in order to parameterize the saturated  hydraulic conductivity and porosity (including the proportion of matrix- and macro-pores) of the bulk waste-rock material. The first field-permeameter was constructed by Neuner (2009). The second permeameter was constructed by Momeyer (in progress) and is described here, since repeat measurements were done and included in Section 2.4 and Appendix C. Figure 29 shows some design features of the field-permeameter prior to it being filled with wasterock. The field-permeameter is 32 m3 (4 m by 4 m by 2 m high). The base of the permeameter was graded using sand and the entire interior covered in an HDPE impermeable liner. A screened PVC drain was sealed to the centre of the basal liner and covered in protective gravel crush. The drain-line was extended horizontally out below the permeameter to facilitate flow measurement during draindown and to allow the permeameter to be filled with water from the base up during porosity and constant-head tests. Three manometers were installed around the inside edges of the permeameter and screened above the gravel crush in order to measure water level and calculate the hydraulic gradient across the permeameter. An upper drain was installed by cutting a hole in the wall at the top of the permeameter in order to measure outflow during constant-head tests. The permeameter was then filled with NAG, Type I run-of-mine material. A maximum grain-size of 0.5 m3 (0.8 m x 0.8 m x 0.8 m) was  19  used in accordance with Neuner (2009), who used the concept of a representative elemental volume (REV) to determine the largest grain-size that would result in a 1% reduction in porosity of a 16 m3 sample if an additional largest grain-size clast were added. The wasterock at DDMI includes fragments up to 3 m3, resulting in an REV for porosity of 80 m3 (Neuner, 2009). Figure 2-10 illustrates the materials and equipment set-up required for the permeameter experiment. Two water storage tanks (1,2) were filled with raw water from Lac de Gras. The main storage tank (1) was graduated in order to measure the volume of water entering the permeameter. Water was pumped from (1) to the constant head reservoir (3), which has three overflow ports leading back to (1). Multiple overflow ports at different elevations allowed for different constant heads to be applied. Water from (3) was directed down to the lower drain-pipe (6), allowing the permeameter to be saturated from the base up. Manometers (4) are located around the edges of the permeameter. The upper drain (5) was fitted with a T-intersection, allowing overflow from the permeameter to be routed back to a storage tank or allowing outflow to be measured directly at the drain when required. The extra water tank (2) was used to replenish the water in the main storage tank when needed.  2.3  Instrumentation This section summarizes the instrumentation of the Test Piles Research Area as it  pertains to hydrology. For a more in-depth description of instrument construction, readers are directed to Neuner (2009). For a comprehensive documentation of all test piles instrumentation, readers are directed to Smith (2009).  20  2.3.1  Meteorological station DDMI operates a meteorological station approximately 1 km from the Test Piles  Research Area. The station is 440 m AMSL, while the crowns of the test piles are approximately 452 m AMSL. The meteorological station measures and records air temperature, relative humidity, wind speed, wind direction, and shortwave solar radiation 2 m above ground level. Hourly data from 2006 through 2011 was supplied by the Environment department at DDMI for use in this thesis.  2.3.2  Rain gauge tipping buckets Rainfall was recorded at the Test Piles Research Area. Young Model 2202 tipping  bucket rain gauges were installed at the crown of each test pile, approximately 2 m above the waste-rock surface. Figure 2-11 shows a rain gauge tipping bucket installed on top of the instrument hut at the Type I test pile crown. There are a total of four tipping bucket rain gauges: one at the Type I test pile, one at the Type III test pile, and two at the Covered test pile.  2.3.3  Tensiometers Tensiometers were used at the Test Piles Research Area to measure the matric  potential (soil tension/suction) of the fine grained fraction of the waste-rock during thawed conditions. Shallow (0.3 m to 1.2 m) boreholes were drilled into the waste-rock. The tensiometers were placed into the boreholes, which were then backfilled with fine grained material. Model 2725ARLNG tensiometers were used, with jet fill reservoirs to facilitate water addition, and current transducers (Model 5301-B.5) to enable automated data  21  collection every 0.5 hours (purchased from Soilmoisture Equipment Corp., Santa Barbara,CA). No tensiometers were installed at the Type I test pile. Four tensiometers were installed on the Type III test pile crown; two tensiometers measure tension at a depth of 0.6 m and the other two at a depth of 1.2 m. Four tensiometers were installed on the Covered test pile crown within the Type I cover; two tensiometers measure tension at a depth of 0.6 m and the other two at a depth of 1.2 m. Three tensiometers were installed adjacent to the AZLs; the tensiometers record tension at a depth of 0.3 m, 0.6 m, and 0.9 m respectively.  2.3.4  2.3.4.1  Soil moisture sensors  ECH2O sensors Decagon Devices ECH2O TE and 5TE soil moisture sensors were used at the Test  Piles Research Area to estimate water content and temperature of the fine grained fraction of the waste-rock. The 5TE has replaced the ECH2O‎TE‎in‎Decagon‎Devices’‎soil‎moisture‎ sensors line-up, but both probes measure water content, temperature, and electrical conductivity. Using a sensor that measures both moisture content and temperature at one location simultaneously is extremely useful in a northern climate like Diavik, where the waste-rock undergoes annual freezing and thawing. The sensors were calibrated in the <5 mm fraction of the waste-rock material used to construct the test piles, yielding a calibration curve accurate to within 0.02 m3/m3 (Neuner, 2009). Shallow boreholes were first drilled into the waste-rock and then backfilled with fine grained material as the sensors were installed at depth. Decagon Devices Em50 data loggers were used to power the sensors and store data.  22  Decagon Devices soil moisture sensors were installed at the AZLs and the Covered test pile. At the AZLs three sensors were installed adjacent to the tensiometers (20 cm away) at 0.3 m, 0.6 m, and 0.9 m depths in order to compare matric potential and volumetric moisture content of the material at similar depths through time. At the Covered test pile the surviving sensors include two sensors installed within the Type III material and till cover at the crown of the pile, three sensors at the top of the till cover (two on the east slope-face and one on the west slope-face), and two at the bottom of the till cover (one on the west slopeface and one on the south slope-face).  2.3.4.2  Time domain reflectrometry (TDR) sensors Time domain reflectrometry (TDR) sensors were installed at the Test Piles Research  Area in order to measure moisture content at various horizontal and vertical locations extending below the crowns of the test piles. Figure 2-12 shows an example of the sensors constructed by the project, following the design of Nichol et al. (2002). Prior to installation, the sensors were placed in a 15 cm diameter permeable nylon fabric sock filled with 1 cmminus material and calibrated in waste-rock material with particle size <20 mm. Neuner (2009) found that the electromagnetic field generated by the TDR sensors extended outside of the matrix material in the permeable sock. Probes surrounded by material with cobbles and air-filled pores measured moisture contents that deviated from the calibration curve developed using only matrix and cobbles. This finding is relevant to measurement at the test piles, since large air-filled voids exist in the heterogeneous material. Neuner (2009) estimated that the uncertainty in volumetric moisture content resulting from the calibration curve is 0.05 m3/m3, while the uncertainty in volumetric moisture content resulting from the  23  application of the calibration curve to sensors surrounded by large rocks and/or air-filled pores could result in an underestimation of moisture content inside the nylon sock by as much as 0.07 m3/m3. All TDR sensors were installed below the crown of each test pile, with none installed within the batters. The probes were placed, in their nylon socks, along the tip faces of the test piles as they were constructed. Figures 2-13 to 2-15 are cross-sections of the test piles showing the locations of the TDR sensors. Only the sensors that survived the construction phase are shown. Four of ten sensors survived in the Type I pile, thirteen of twenty-eight sensors survived in the Type III pile, and ten of eleven sensors survived in the Covered pile. Over the course of the project, from 2006 through 2011, only one TDR sensor has failed (33N2tdr01 failed in 2008). The naming convention used to label the tensiometers uses a string‎nine‎characters‎long.‎The‎first‎character‎is‎for‎the‎test‎pile‎(e.g.‎‘1’‎means‎Type‎I‎test‎ pile),‎the‎second‎character‎is‎the‎tip‎face‎(e.g.‎‘2’‎means‎tip‎face‎2),‎the‎third‎and‎fourth‎ characters describe the sensors location with respect to the centre line of the test pile (e.g. ‘W2’‎means‎2‎m‎west‎of‎the‎centre‎line),‎the‎fifth through seventh characters indicate the instrument‎(i.e.‎‘tdr’‎in‎this‎case),‎and‎the‎eighth‎and‎ninth‎characters‎indicate‎the‎depth‎of‎ the‎sensor‎below‎the‎test‎pile‎crown‎(e.g.‎‘05’‎means‎5‎m‎below‎the‎crown).‎Table‎2-2 lists all surviving sensors, with their datalogger ID in parentheses.  2.3.4.3  Comments on moisture content and matric potential sensors The ECH2O TE and 5TE sensors, TDR sensors, and tensiometers all facilitate the  monitoring of fluid flow within the test piles. These instruments all have limitations within waste-rock piles however, since they must be in hydraulic contact with fine grained matrix  24  material to function properly. This is problematic when considering that the grain-size distribution of waste-rock material and the construction methods of stockpiles can result in large void spaces and preferential, matrix-free pathways (Section 1.4). This restriction of the instruments highlights the difficulty of monitoring fluid fluxes in waste-rock piles, since flow through matrix-free zones cannot be directly observed. TDR sensors can be used to determine the dielectric permittivity of the matrixfraction of waste-rock. The dielectric permittivity of porous media changes with water content and can be correlated to volumetric moisture content via the Topp et al. (1980) equation and lab calibrations. The dielectric permittivity of the various phases of water are different; meaning that for the test piles, which undergo annual freeze-thaw, dielectric permittivity can only be correlated to moisture content during thawed conditions when water is in liquid phase. In addition, the dielectric permittivity of ice and air are very similar. This means that care has to be taken when making observations regarding whether a location is frozen or dry (Chapter 4).  2.3.5  Lysimeter locations The main basal lysimeter (a.k.a basal drain) and the BCLs at each test pile report to  individual drainpipes that route water out of the test piles to individual flow-cell systems and tipping buckets. Figures 2-16 to 2-18 show the relative locations of the BCLs at the base of the Type I, Type III, and Covered test pile respectively. The main basal drain and BCLs are labeled using a nine character string. Table 2-3 lists the names of the main basal drains and BCLs at each test pile. The first‎character‎indicates‎the‎test‎pile‎(e.g.‎‘1’‎means‎Type‎I‎test‎ pile),‎the‎second‎character‎denotes‎that‎the‎location‎is‎at‎the‎base‎of‎the‎test‎pile‎(i.e.‎‘B’‎for‎  25  base), the third and fourth characters describe the relative location of the lysimeter within the test‎pile‎(e.g.‎‘WB’‎means‎west‎batter,‎‘NC’‎means‎north‎centre,‎‘xx’‎is‎a‎place‎holder‎for‎the‎ main basal drains), the fifth through seventh characters differentiate basal liner versus BCL (i.e.‎‘drn’‎for‎basal‎drain,‎‘lys’‎for‎BCL),‎and‎the‎eighth and ninth characters indicate the depth‎of‎the‎basal‎drain‎(e.g.‎‘15’‎means‎15‎m‎below‎the‎crown)‎or‎the‎dimensions‎of‎the‎ BCL‎and‎secondary‎location‎(e.g.‎‘2W’‎means‎the‎west-most 2m x 2m BCL).  2.3.6  Flow cells Effluent originating from the main basal drains and BCLs is directed out of the test  piles and into a flow-cell system for geochemical data collection. Figure 2-19 shows the flow-cell set-up at the Covered pile basal drain (CBxxdrn13). The set-up consists of a sampling cell for manual geochemical sampling, a pH cell for automated pH and temperature measurement, and an EC cell for automated electrical conductivity measurement. Effluent flows first into the sampling cell, then to the pH cell, then to the EC cell, and finally to a tipping bucket.  2.3.7  Tipping buckets Tipping buckets were used at the Test Piles Research Area to measure outflow rates  and volumes from the main basal drain of each test pile, the BCLs, and the AZLs. Custom tipping buckets, fabricated at the University of British Columbia, were used for the main basal drains. Figure 2-20 shows an example of a custom tipping bucket. Young Model 2202 tipping bucket rain gauges were used for all BCLs and AZLs. Figure 2-21 shows an example of the tipping bucket rain gauges at the Type III BCL outflow. All tipping buckets were  26  calibrated to determine the non-linear relationship between tip-time and flow rate. Appendix B details the design of the custom made tipping buckets and documents the tipping bucket calibration equations.  2.3.8  Dataloggers and telemetry The project uses Campbell Scientific CR10X and CR1000 loggers, as well as  Decagon Em50 loggers, for automated collection of all hydrology data. A telemetry system at the Test Piles Research Area transfers all data from the dataloggers to an on-site computer at scheduled intervals.  2.4  Key values from previous work This section summarizes key waste-rock properties reported by Neuner (2009) and  Neuner et al. (2012), who conducted a comprehensive set of lab-scale experiments to characterize the hydrogeologic properties of the matrix-fraction of the run-of-mine material used in the construction of the test piles. A field-scale experiment was also conducted (Neuner, 2009; Momeyer, in progress; Fretz, this thesis) to determine the saturated hydraulic conductivity and porosity of a bulk waste-rock sample, and a field-scale grain-size analysis was carried out to determine the grain-size distribution of the bulk waste-rock. The results summarized below are instrumental to the analysis contained within this thesis. Readers are directed to Neuner (2009) and Neuner et al. (2012) for additional methods and discussion regarding these results.  27  2.4.1  Bulk waste-rock properties Bulk waste-rock refers to run-of-mine material consisting of all grain size fractions.  Hydraulic properties of the bulk wastes-rock include the influence of all pore space (i.e. matrix pores and larger voids) within the stockpile.  2.4.1.1  Grain-size distribution The grain-size distributions of the Type I and Type III waste-rock that compose the  test piles are very similar. Figure 2-22 is a composite grain-size distribution from Neuner et al. (2012) consisting of a complete 90 t sample distribution and two 260 t sample distributions that consist of only the boulder fraction of the waste-rock. The matrix-fraction (<5 mm diameter grains) makes up 18% of the bulk waste-rock by volume, with the remaining fraction consisting of larger grain-sizes. Chi (2011) photographically analyzed the grain-size distribution visible on the Type I and Type III test pile batters. A computer program was developed to characterize grain-size distribution from photographic images (for grain-sizes >0.1 m). Results indicated that the grain-size increases non-linearly from the top of the test piles to the bottom, with the finer fraction retained near the top. Additionally, it was determined that the >0.1 m grain-size fraction constituted the greatest mass, but the finer <0.1 m fraction constituted the greatest surface area.  2.4.1.2  Hydraulic properties Neuner et al. (2012) estimated the porosity and saturated hydraulic conductivity (Ksat)  of the bulk waste-rock using a 16 m3 field-permeameter filled with run-of-mine waste-rock.  28  Momeyer (in progress) and Fretz (this thesis) also conducted similar experiments using a 32 m3 field-permeameter filled with run-of-mine waste-rock. Results from Fretz (this thesis) are summarized here for comparative purposes, but are described in detail in Appendix C. Neuner et al. (2012) and Momeyer (in progress) estimated bulk porosity to be 0.24 and 0.27 respectively. Drain-down tests were conducted at the field-permeameters to estimate the fraction of porosity attributed to matrix-pores and macro-pores. Atmospheric conditions were induced at the base of the permeameters by opening the basal drains. Outflow from the fieldpermeameters and hydraulic gradient across the field-permeameters were measured during drain-down. It was assumed that the volume of outflow recorded from fully-saturated conditions to the onset of unsaturated conditions at the base approximates the volume of pore-space in the macropores (i.e. the fraction with negligible capillarity). Conversely, the volume of water remaining at the onset of unsaturated conditions at the base corresponds to the volume of pore-space in the matrix fraction (i.e. the fraction with capillarity). Manometers screened at the base of the permeameters allowed the onset of unsaturated conditions to be observed. Neuner et al. (2012), Momeyer (in progress), and Fretz (this thesis) estimated the porosity attributable to macropores to be 0.18, 0.22, and 0.19 respectively. Likewise, the porosity attributable to the matrix material was estimated to be 0.07, 0.05, and 0.08. An alternative estimate of the porosity attributable to the matrix fraction is achieved by multiplying the porosity of the matrix material (0.25) by the matrix fraction of the bulk waste-rock (18%), yielding 0.05. As a word of caution, these values cannot be used to comment on the predominance of matrix versus preferential flow. Constant head tests conducted by Neuner et al. (2012), Momeyer (in progress), and Fretz (this thesis) yielded estimates for Ksat of 1 x 10-2 m/s, 4 x 10-3 m/s, and 6 x 10-3 m/s  29  respectively. Recall, the estimate from Neuner et al. (2012) was from a different fieldpermeameter than the one used in Momeyer (in progress) and Fretz (this thesis). Table 2-4 summarizes the porosity and Ksat values obtained at the field-permeameters. Appendix C provides additional details of the field-permeameter tests and the uncertainty in the above estimates.  2.4.2  2.4.2.1  Matrix-fraction (<5 mm) properties  Grain-size distribution 197 hand samples of material <100 mm were collected at the Type I test pile (91  samples), the Type III test pile (101 samples), and the AZLs (5 samples). Test pile samples were collected from the 2m basal lifts and each tip face. Sieving of the hand samples indicated that the <5 mm matrix fraction of the bulk waste-rock (18% by volume) consists of 92% sand and 8% silt and clay, on average (Neuner et al., 2012).  2.4.2.2  Hydraulic properties Gravimetric measurement of porosity yielded an average value of 0.25 for the matrix  material, with a range of 0.23 to 0.27 (Neuner et al., 2012). Lab-permeameter constant head tests conducted by Neuner et al. (2012) yielded Ksat estimates with a range of 2 x 10-6 m/s to 3 x 10-5 m/s and a geometric mean of 9 x 10-6 m/s. The above tests were conducted on 18 of the sieved hand samples from the Type I test pile, Type III test pile, and AZLs. Three trials were carried out for each of the 18 samples.  30  Wetting retention curves (WRCs) of the matrix material were developed by Neuner et al. (2012) using two methods. The first method involved lab-based measurement using Tempe cells. Five samples of Type III matrix material were used. Drying curves were constructed and yielded van Genuchten‎(1980)‎shape‎parameters‎as‎follows:‎an‎α‎range‎of‎0.1‎ kPa-1 to 0.9 kPa-1, an n range‎of‎1.4‎to‎2.5,‎a‎saturated‎moisture‎content‎(θsat) range of 0.16 to 0.24,‎and‎a‎residual‎saturation‎(θres) range of 0.005 to 0.05. The second method involved an in situ, field-based measurement using moisture content data and tension data from the ECH2O sensors and tensiometers adjacent to the AZLs. In this case, a wetting curve was constructed using data collected over a 60 hour period as a wetting front from an applied rainfall event was recorded at the sensors. The wetting curve van Genuchten (1980) shape parameters‎are‎as‎follows:‎an‎α‎of‎0.9‎kPa-1, an n of‎1.7,‎a‎θsat of‎0.22,‎and‎a‎θres of 0.04. These parameters are representative of the matrix material surrounding the ECH2O sensors and tensiometers. Table 2-5 summarizes the above matrix parameters. Figure 2-23 shows the WRCs and hydraulic conductivity curves for the matrix material.  2.4.3  2.4.3.1  Field-scale experiments (test piles and active zone lysimeters)  Field capacity and initial moisture content Neuner et al. (2012) estimated the field capacity of the bulk waste-rock by observing  the volume of net infiltration required to initiate outflow at the Type III AZLs and taking into account the initial moisture content of the AZLs. A bulk field capacity of 0.06 m3/m3 was estimated. The initial moisture content of the bulk waste-rock was estimated by taking the initial moisture content of the <40 mm fraction (0.025 m3/m3) and multiplying it by the  31  fraction of the waste-rock finer than 40 mm (0.35). This yielded a bulk initial moisture content of 0.01 m3/m3.  2.4.3.2  Infiltration capacity Neuner et al. (2012) conducted ring infiltrometer tests at the crown of the Type III  test pile to estimate infiltration capacity of the matrix-supported regions, under thawed conditions. Seven single ring infiltrometers were set up approximately 1 cm into the crown and the outer edges sealed with bentonite. Infiltration capacity had a range of 6 x 10-7 m/s to 2 x 10-5 m/s and a geometric mean of 5 x 10-6 m/s. Differences in infiltration capacity between the crowns of the test piles and the surfaces of the AZLs was qualified by observing the onset of surface ponding. Local surface ponding at the crowns of the test piles was observed for rainfall rates of and exceeding 2 x 10-6 m/s, while surface ponding was not observed at the surfaces of the AZLs for the highest rainfall rate of 3.3 x 10-6 m/s. This is likely attributed to differences in the traffic surface created at the respective locations during construction. Although surface ponding has been observed at the crowns of the test piles, surface runoff does not occur. Instead, small-scale runoff to areas of higher infiltration capacity occurs and facilitates infiltration.  2.5  Concluding statement Chapters 3 and 4, representing the body of this thesis, reference this chapter for  details regarding study site and experimental set-up and design. The summary of previous work included key values necessary for interpreting results in Chapters 3 and 4. Properties of the matrix fraction are employed in the evapotranspiration computation presented in Chapter  32  3, while both matrix and bulk-waste rock properties are used in Chapter 4 to interpret wet-up and fluid flow within the AZLs and test piles.  33  2.6  Figures  Figure ‎2-1  Location of the Diavik Diamond Mine (64 o29’‎N,‎110o18’‎W)  Figure ‎2-2  Diavik Diamond Mine on East Island  34  Figure ‎2-3  Aerial photograph of the Test Piles Research Area  35  Figure ‎2-4  Schematic of the Test Piles Research Area. Red denotes instrument huts at the base of  the piles. Green denotes instrument huts at the crowns of the piles. Re-created and adapted from Smith (2009)  36  Figure ‎2-5  Pre-construction design of the Type I (a), Type III (b) and Covered (c) test piles. Solid  black lines denote the basal extent of each pile. Dotted black lines denote the crowns of each pile. Dashed red lines denote the basal drain line and the arrows denote flow direction. Rectangular cells are locations of basal collection lysimeters. Re-created from Smith (2009)  37  Figure ‎2-6  Post-construction drawings of the Type I test pile and drainage scheme. (a): white  represents the crown, grey represents the batters, and the square patterns represent the basal collection lysimeters. (b): contour interval is 0.2 m. Adapted from Neuner (2009)  38  Figure ‎2-7  Post-construction drawings of the Type III test pile and drainage scheme. (a): white  represents the crown, grey represents the batters, and the square patterns represent the basal collection lysimeters (BCLs). (b): contour interval is 0.2 m. Adapted from Neuner (2009)  39  Figure ‎2-8  During- and post-construction pictures of the active zone lysimeters (AZLs). (a): Empty  water drums and drainlines. From foreground to background the water drums represent the Type III West, Type III East, Type I West, and Type I East AZLs. (b): AZLs being filled with run-of-mine material. (c): surface of the completed AZLs. (a) and (b) sourced from Neuner (2009)  40  Figure ‎2-9  Design of the 32 m3 field-permeameter. Middle image shows the slotted drain at the base  of the field-permeameter. Image source: Momeyer (in progress)  Figure ‎2-10  Field-permeameter - experiment set-up  41  Figure ‎2-11  Rain gauge tipping bucket at the crown of the Type I test pile  42  Figure ‎2-12  Custom-built time domain reflectrometry (TDR) sensors. Image source: Neuner (2009)  Figure ‎2-13  North-South cross-section of the Type I test pile, 2 m west of the centre line. TDR sensor  locations denoted by red dots. Adapted from Smith (2009)  43  Figure ‎2-14  (a): West-East cross-section of the Type III test pile, 2 m south of the centre line.  (b): West-East cross-section of the Type III test pile, 2 m north of the centre line. TDR sensor locations denoted by red dots. Adapted from Smith (2009)  44  Figure ‎2-15  (a): South-North cross-section of the Covered test pile, 2 m east of the centre line.  (b): South-North cross-section of the Covered test pile, 2 m west of the centre line. TDR sensor locations denoted by red dots. Adapted from Smith (2009)  45  Figure ‎2-16  Relative locations of the basal collection lysimeters (BCLs) at the base of the Type I test  pile. Consult figures 2-5 and 2-6 for overall location within the test pile.  Figure ‎2-17  Relative locations of the basal collection lysimeters (BCLs) at the base of the Type III  test pile. Consult figures 2-5 and 2-7 for overall location within the test pile.  Figure ‎2-18  Relative locations of the basal collection lysimeters (BCLs) at the base of the Covered  test pile. Consult figure 2-5 for overall location within the test pile.  46  Figure ‎2-19  Flow-cell set-up at the Covered test pile basal drain outflow (CBxxdrn13)  Figure ‎2-20  Uninstalled custom-built tipping bucket manufactured at UBC  47  Figure ‎2-21  Young Model 2202 tipping bucket raingauges at the Type III BCL outflow  Figure ‎2-22  Grain-size distribution of the run-of-mine waste-rock. Source: Neuner (2009)  48  Figure ‎2-23  (a): Water retention curves of matrix material. (b): Hydraulic conductivity curves of  matrix material. Source: Neuner et al. (2012)  49  2.7  Tables  Table ‎2-1  Test pile and active zone lysimeter (AZL) dimensions  Base Dimensions  Location  Type I Test Pile Type III Test Pile Covered Test Pile Type I AZLs Type III AZLs  Table ‎2-2  Crown Dimensions  Height of Crown Above Drain  Test Piles 50m by 60m 1350m3 50m by 60m 1350m3 80m by 125m ? Active Zone Lysimeters 2 3.6m 3.6m2 2.1m2 2.1m2  Drainage Scheme  13m 15m 13m  Routed thru centre Routed to edges Routed thru centre  1.45m 1.7m  -  Surviving time domain reflectrometry (TDR) sensor labels and datalogger IDs  Type I test pile 12W2tdr06 (TDR80) 12W2tdr09 (TDR 83) 11W2tdr02 (TDR86) 11W2tdr03 (TDR87)  Type III test pile 31N2tdr01 (TDR51) 31N2tdr07 (TDR52) 31N2tdr09 (TDR53) 31S2tdr01 (TDR54) 31S2tdr03 (TDR55) 31S2tdr05 (TDR56) 31S2tdr09 (TDR57) 32N2tdr07 (TDR58) 33N2tdr01 (TDR59)* 33N2tdr03 (TDR60) 33N2tdr05 (TDR61) 33S2tdr02 (TDR62) 33S2tdr03 (TDR63)  Covered test pile C2E2tdr06 (TDR90) C2E2tdr07 (TDR91) C2E2tdr08 (TDR92) C2E2tdr10 (TDR93) C2W2tdr06 (TDR94) C2W2tdr07 (TDR95) C2W2tdr08 (TDR96) C3E2tdr06 (TDR98) C3E2tdr08 (TDR99) C3E2tdr10 (TDR100)  *Sensor failed in 2008  50  Table ‎2-3  Test pile basal drain and basal collection lysimeter labels and datalogger IDs  Location Type I basal drain Type III North basal drain Type III South basal drain Covered pile basal drain Type I BCLs  Type III BCLs  Table ‎2-4  Label 1Bxxdrn13 3BNxdrn15 3BSxdrn15 CBxxdrn13 1BWBlysCluster 1BWClysCluster 1BEClysCluster 3BNBlys2W 3BNBlys2E 3BNBlys4W 3BNBlys4E 3BNClys2W 3BNClys2E 3BNClys4W 3BNClys4E 3BSClys2W 3BSClys2E 3BSClys4W 3BSClys4E  Datalogger ID tb_13 tb_14 tb_15 ? tb_10 tb_11 tb_10 tb_1 tb_1 tb_2 tb_3 tb_4 tb_4 tb_5 tb_6 tb_7 tb_7 tb_8 tb_9  Bulk waste-rock properties – Field-permeameter experiments  Experiment  Permeamete r Volume (m3) 16  Bulk Porosity  0.24 Neuner (2009) Momeyer (in 32 0.27 progress) 32 0.27* Fretz (this thesis) *Value adopted from Momeyer (in progress)  Porosity Attrib. to Macro-pores 0.18  Porosity Attrib. to Matrix  Ksat (m/s)  0.07  1 x 10-2  0.22  0.05  4 x 10-3  0.19  0.08  6 x 10-3  51  Table ‎2-5  Ksat (geometric mean) (m/s) 9x10-6  Table ‎2-6  Matrix material hydraulic properties  Ksat (Range) (m/s) 2x10-6 to 3x105  Porosity 0.23 to 0.27  α‎(kPa-1)  n  0.1 to 0.9 1.4 to 2.5  θsat  θres  0.16 to 0.24  0.005 to 0.05  Miscellaneous field-scale experiment parameters  Bulk field capacity at AZLs (m3/m3) 0.06 TIII AZLs (Neuner) 0.04 to 0.06 TI AZLs (Fretz)  Infiltration capacity at Type III test pile crown (m/s) -7 -5 6 x 10 to 2 x 10 Range -6 5 x 10 Geometric mean  52  Chapter 3: Infiltration into waste-rock in a northern climate  3.1  Introduction Estimates of net infiltration are required to understand the character and distribution  of solute migration within a waste-rock stockpile. Consequently, understanding net infiltration is an important component of mine closure planning. The objectives of this chapter are to 1) estimate net infiltration into the active-zone lysimeters (AZLs) and crowns of the uncovered test piles (crowns of the UTPs) and 2) evaluate the applicability of an indirect evapotranspiration computation for prediction of net infiltration using meteorological data and surface properties of waste-rock piles. Estimating net infiltration requires an estimate of the water balance components. Equation 3-1 is a standard water balance equation, 3-1  where P is precipitation, ET is evapotranspiration, RO is runoff, and ΔS is change in storage over the time-frame in question. Quantifying the evapotranspiration component is challenging and there exist multiple methods for doing so, some based on direct measurement, and others based on calculation using meteorological data and correction factors. The main direct methods for measuring ET include energy balance, microclimatological, soil water balance, and lysimetry methods. Energy balance methods involve estimating or measuring energy fluxes towards and away from the ground surface. The latent heat flux, which represents the evapotranspiration component, can be calculated from a conservation of energy equation if all other components are known (i.e. net radiation,  53  sensible heat, and ground heat flux). Microclimatological methods use a mass transfer model to estimate evapotranspiration by measuring wind speed and water vapour gradients above a surface. The eddy covariance technique is one such method, and uses measurements of wind speed and gas concentrations to calculate turbulent fluxes of latent and sensible heat above a surface (Burba and Anderson, 2010). Soil water balance methods involve the measurement of water flux components in a water balance (e.g. Equation 3-1); ET can be calculated if all other fluxes in the water balance are known. Lysimetry methods involve the sequential weighing of precision lysimeters to determine the mass of water lost to evapotranspiration (Howell, 1991), or the measurement of drainage at the base of the lysimeter to determine evapotranspiration from a known input of water (Gee and Hillel, 2006). Direct methods of evapotranspiration are generally research oriented, since the methods commonly require expensive equipment, certain expertise to conduct, and are non-ideal for repeat measurements. However, direct measurements are excellent for the comparison and validation of indirect, computational methods. The main methods based on computation are semi-empirical and utilize meteorological data to estimate a potential evapotranspiration (PET) or reference evapotranspiration (ETo), which can then be adjusted to actual evapotranspiration using coefficients based on crop and material surface properties or other correction factors. PET refers to the evapotranspiration that would occur at a surface that is not short of water, and therefore not restricted by water availability. ETo refers to evapotranspiration from a standard reference surface with specific properties, also not limited by water availability. There are various formulations for PET and ETo. Examples of methods for calculating PET include formulations by Thornthwaite and Holzman (1939), Penman (1948), Monteith (1965), and  54  Granger and Gray (1989). Two common methods for calculating ETo include the Hargreaves et al. (1985) formulation and the FAO recommended Penman-Monteith formulation (Allen et al., 1998). PET and ETo can also be estimated by first measuring evaporation from an open body of water using pan evaporation techniques (Hydrological services, 2003) and then employing empirical correction factors (e.g. Grismer et al., 2002).  3.1.1  ET studies in waste-rock Actual evapotranspiration (actual evaporation for bare waste-rock) is seldom  measured directly in mining applications; common mine closure and restoration practices involve the estimation of potential evapotranspiration, using evaporation pans or semiempirical atmospheric computations, and the subsequent estimation of actual evaporation using correction factors. In the literature, a common method of estimating evaporation from waste-rock is via numerical simulations that take into account heat and mass transfer processes. Numerical simulation studies have mostly focused on the evaluation of soil covers that are used to reduce infiltration of water into potentially reactive waste-rock (e.g. Fayer et al., 1992; Wilson et al., 1994; Yanful and Choo, 1997; Yanful et al., 2006). In research settings, actual evaporation has been measured directly from tailings and waste-rock using microclimatological techniques (e.g. Fujiyasu et al., 2000; Newson and Fahey, 2003; Carey et al., 2005) and lsyimetry techniques (e.g. O`Kane et al., 1998; Fujiyasu et al., 2000). Here, the FAO-56 Penman-Monteith (FAO-PM) method is used to estimate reference evapotranspiration at the waste-rock surface and then a dual crop coefficient method to estimate actual evaporation and net infiltration (Allen et al., 1998; described below). Neuner et al. (2012) used a similar method to estimate reference evapotranspiration from a waste-  55  rock surface; however, actual evaporation and net infiltration were calculated using an alternate method. Here, net infiltration calculated by the FAO-PM method is compared to direct measurement of net infiltration at lysimeters within the upper 2 m of waste-rock, in order to evaluate the applicability of the FAO-PM method to waste-rock. Carey et al. (2005) compared direct estimates of actual evaporation, calculated via eddy covariance, to an indirect estimate of actual evaporation using the Granger and Gray (1989) modified Penman method. A significant conclusion of that study was that a basic set of atmospheric data could be used to estimate actual evaporation from a mine waste-rock pile with reasonable accuracy. The estimation of net infiltration into the AZLs and crowns of the UTPs will: 1) highlight periods and conditions that are most important for net infiltration into waste-rock in a northern climate (this Chapter), 2) aid in evaluating the applicability of the FAO-PM method to waste-rock (this Chapter), 3) assist in the analysis of fluid flow through the test piles and AZLs (Chapter 4), 4) assist in evaluating the observed hydrologic differences between the Type I and Type III uncovered test piles (Chapter 4), and 5) assist in closing a water balance for the test piles in order to understand controls on the temporal and spatial variations in solute loading (future work).  56  3.2  Methods  3.2.1  Study site and meteorological instrumentation A complete description of the study site was given in Chapter 2; Sections 2.1 and 2.2  provided an overview of DDMI and the Test Piles Research Area. Rainfall instrumentation was documented in Section 2.3.2 and details of the meteorological station were documented in Section 2.3.1.  3.2.2  Active zone lysimeter (AZLs) water balance Section 2.2.2 outlined the construction, design, and instrumentation of the AZLs.  AZL locations are referred to as the TIE (Type I East), TIW (Type I West), TIIIE (Type III East), and TIIIW (Type III West) AZLs in the figures. A water balance has been used to determine annual net infiltration at the AZLs, which is much easier to compute than a water balance for the uncovered test piles. The AZLs have a simple geometry, no batters, and are less than 2 m in height. Thus, the small scale of the AZLs enabled changes in storage to be more accurately observed with in situ instrumentation. This is in contrast to the uncovered test piles, which are large enough and geometrically complex enough that changes in water storage over a year were difficult to monitor with the in situ instrumentation. The water balance for the AZLs was simplified from Equation 3-1: 3-2  where RF is rainfall and E is evaporation. The following simplifications were made:   Precipitation was simplified to rainfall, in order to exclude infiltration from snowmelt. Observations from 2007 through 2011 indicated that wind scoured a majority of the snow from the upper surface of the AZLs prior to the onset of thawing 57  conditions, leading to minimal (considered negligible here) infiltration of snowmelt each year.   Runoff has been set to zero, following observations of no runoff at the AZLs (Section 2.4.3.2).    The transpirative term was removed since there is no vegetative cover at the upper surface of the AZLs.    Water balances were only calculated for years after the AZLs had initially wet-up. Post wet-up, ECH2O sensors at the AZLs indicated negligible changes in storage on a yearly basis. Therefore, post wet-up, annual net infiltration is approximately equal to annual outflow  The crowns of the UTPs share several of the conditions listed above. Wind scours a majority of the snow from the crowns of the UTPs prior to the onset of thawing conditions. Figures 3-1 and 3-2 provide examples of this in 2012. In addition, there is no vegetative cover at the crowns of the UTPs, or observation of large-scale runoff from the crowns of the UTPs (Section 2.4.3.2). A qualitative comparison of surface texture (Figure 3-3) shows that the upper surface of the AZLs covers the range in surface texture at the crowns of the UTPs. The crowns of the UTPs are almost entirely composed of regions representative of the Type I East, Type III East, and Type III West AZLs, which have significant matrix material at their upper surfaces; however, infrequent cobble regions representative of the Type I West AZL surface also exist at the crowns of the UTPs. The surfaces of the AZLs are directly beside, and at the same elevation as, the crowns of the UTPs and are, therefore, subject to the same climatic conditions.  58  3.2.3  Moisture content and matric potential sensors Changes in near-surface moisture contents below the crowns of the UTPs are  presented as a qualitative indicator of net infiltration. Figures 2-13 and 2-14 showed the locations of the individual TDR probes; a probe 2 m below the Type I test pile crest (11W2 2m) and a probe 1 m below the Type III test pile crest (31S2 1m) are used here. These locations have been chosen because they are the shallowest TDR sensors within these respective piles and because they are at similar depths to the AZL drains. Fluid fluxes, calculated by Neuner et al. (2012) past two tensiometer profiles located at the Type III test pile crown are presented for comparison to the calculated FAO-PM net infiltration (described below). Each tensiometer profile (North and South) consists of a tensiometer at 0.6 m and 1.2 m depths. Additional information regarding TDR sensors and tensiometers was given in Sections 2.3.3 to 2.3.4.3.  3.2.4  FAO-56 Penman-Monteith method The FAO-56 Penman-Monteith method (FAO-PM) was used to calculate evaporation  and net infiltration at the AZLs and crowns of the UTPs. Evaporation (Ea(t), actual evaporation as a function of time) was calculated by first estimating ETo(t) using the FAO-56 Penman-Monteith formulation (Allen et al., 1998 chapter 4) and then employing a dual crop coefficient to estimate Ea(t) (Allen et al. 1998 chapter 7). The dual crop coefficient method calculates water loss from a soil evaporation layer in two ways: 1) evaporation (Ea(t)) upwards and 2) deep percolation downwards. Here, removal of water from the zone of evaporative influence via deep percolation is referred to as net infiltration. These steps are outlined in detail below.  59  3.2.4.1  Estimating reference evapotranspiration ETo(t) describes the evapotranspiration rate from a hypothetical grass reference crop  that is not short of water. ETo(t) was calculated on an hourly basis (Equation 3-3) and summed daily. An hourly time step was chosen, as opposed to a daily time step, because hourly time steps are more representative of ETo(t) in areas where wind speed and cloud cover change substantially during a day. Fortunately, DDMI collects hourly meteorological data. ( ( )  )  ( (  )  (  (  )  )  )  3-3  ETo(t) is standardized reference evapotranspiration [mm h-1], Rn is net radiation at the crop surface [MJ m-2 h-1], G is soil heat flux density [MJ m-2 h-1], Thr is mean hourly air temperature [oC],‎∆‎is‎saturation‎slope‎vapour‎pressure‎curve‎at‎Thr [kPa oC-1],‎γ‎is‎the‎ psychrometric constant [kPa oC-1], eo(Thr) is saturation vapour pressure at air temperature Thr [kPa], ea is average hourly actual vapour pressure [kPa], and u2 is average hourly wind speed [m s-1]. The following assumptions and approximations have been made in the calculation of ETo(t):   No hourly atmospheric pressure readings were taken at the Diavik meteorological station. As a result, atmospheric pressure was estimated using Equation 7 in Allen et al. (1998), which is based on the pressure exerted by the weight of the earth`s atmosphere at 20oC. Allen et al. (1998) note that taking an average pressure value for a location is sufficient, since the effect of changing atmospheric pressure on the psychrometric constant is small.  60    Equation 37 in Allen et al. (1998) was used to calculate the incoming shortwave radiation that would reach the surface under cloudless conditions (Rso). The equation uses extraterrestrial radiation and elevation above sea level to estimate Rso, which is necessary to calculate net radiation.    A soil heat flux density (G) of 0.3Rn has been adopted from Carey et al. (2005), which falls within the range suggested by Allen et al (1998) and is comparable to calculations by Pham (2012).    Negative values of hourly ETo, which can occur when longwave radiation from the surface is large during winter in northern latitudes, have been set to zero following work by Choudhury (1997)  3.2.4.2  Estimating actual evaporation A dual crop coefficient method was used to express the difference between ETo(t) and  Ea(t). The dual crop coefficient method splits the single crop coefficient (Kc) into two coefficients: one for crop transpiration (Kcb) and one for soil evaporation (Ke). The formula for crop evapotranspiration is outlined in Equation 3-4, ( )  (  )  3-4  where ETc(t) is crop evapotranspiration [mm d-1], Kcb is the basal crop coefficient, Ke is the soil evaporation coefficient, and ETo is reference evapotranspiration [mm d-1]. Equation 3-4 was modified to Equation 3-5 in order to represent the AZLs and crowns of the UTPs. ( )  3-5  In Equation 3-5, Ea is actual evaporation [mm d-1] and Kf is the frozen soil coefficient [0 or 1]. In the absence of vegetation, Kcb is set to zero and represents the waste-rock surface as bare soil. Therefore, ETc(t) simplifies to Ea(t). Kf has been adopted from Neuner et al. (2012) 61  to limit evaporation during frozen surface conditions. Kf is set to 0 during frozen surface conditions and set to 1 during above-freezing surface conditions. Ke limits evaporation based on the availability of water in an evaporation layer near the soil surface. Ke requires an estimation of: 1) the depth of the evaporation layer that is subject to drying by way of evaporation (Ze), 2) the total evaporable water (TEW; the maximum depth of water that can be evaporated from the completely wetted evaporation layer), and 3) the readily evaporable water (REW; the maximum depth of water that can be evaporated from the evaporation layer without water availability restricting rates). The value of Ke changes with water availability in the soil evaporation layer. Ke is maximal during the energy-limiting stage, when REW has not been fully removed from the evaporation layer. Ke decreases during the falling-rate stage, once REW has been fully removed, and reaches zero once TEW is fully removed. The benefit of using Ke is that it sets evaporation to zero when water availability in the soil evaporation layer is zero. This is important during long periods of no rainfall (common at DDMI) to preserve the water balance in the evaporation layer (Allen et al. 1998). TEW is a function of the field capacity, wilting point, and Ze of the soil or, in this case, waste-rock material. Average field capacity and wilting point of the matrix material (<5 mm waste-rock fraction), calculated by Neuner (2009), were used for the following reasons: 1) The dual crop coefficient method was developed for soils with capillarity, a property that is not exhibited in the waste-rock fraction greater than 5 mm, and 2) the majority of the AZL upper surface (evaporation layer) and crowns of the UTPs are matrix dominated. REW and Ze were determined by calibrating the FAO-PM net infiltration results to the net infiltration calculated via the AZL water balance and to the flux past tensiometer profiles  62  at the Type III test pile. The parameters were adjusted such that the closest match between FAO-PM, AZL water balance, and tensiometer flux net infiltrations was achieved for all years without changing the parameter values between years. A sensitivity analysis using these parameters is given in Section 3.3.3. Table 3-1 documents the Ke parameters described above. The following assumptions and approximations have been made in the calculation of Ea(t):   Runoff has been set to zero for each rain event, corresponding to visual observations at the AZLs and crowns of the UTPs (Section 2.4.3.2).    At the onset of freezing conditions each year, the water remaining in the evaporation layer‎is‎carried‎over‎to‎the‎next‎year’s‎analysis‎to‎represent‎water‎that‎has‎been‎locked‎ in place over the winter.  3.2.4.3  Estimating net infiltration In the dual crop coefficient method, deep percolation occurs when the water content  in the evaporation layer exceeds field capacity following a rain event. Deep percolation was calculated as part of a daily water balance (Equation 3-6), (  )  3-6  where DPe,i is deep percolation loss from the evaporation layer on day i [mm], Pi-ROi is precipitation minus runoff on day i [mm], and De,i-1 is cumulative depth of evaporation from the evaporation layer at the end of day i-1 [mm]. Equation 3-6 is written to show that if the water content in the evaporation layer is below field capacity (De,i-1 > 0), no deep percolation  63  occurs (DPe,i = 0). The cumulative deep percolation over a year corresponds to annual net infiltration.  3.3  3.3.1  Results and discussion  Climate At a latitude of 64o29’‎N,‎DDMI‎receives‎incoming‎solar‎radiation‎that‎varies‎  substantially throughout the year. Incoming solar radiation is highest in May, June, and July when daylight hours are at a maximum and significantly lower September through March when daylight hours are fewer (Figure 3-4). Air temperatures also vary substantially throughout the year. A baseline study using data from regional stations (some having records greater than 50 years) yielded a mean annual air temperature of -10.1 oC for the region (Golder Associates, 2008). Over a five year period, from 2007 through 2011, mean annual air temperature was -8.8oC at the test piles. Figure 3-5 shows air temperature for 2010 (all years documented in Appendix A). Above-freezing conditions typically occur from May to October and influence the development of the active-zone within the test piles and AZLs (Chapter 4), as well as the form of precipitation reaching the waste-rock surface (i.e. rain versus snow). The same baseline study yielded a mean annual precipitation for the region of 351 mm of water equivalent, with 164 mm (47%) as rain and 187 mm (53%) as snow (Golder Associates, 2008). Over a five year period, from 2007 through 2011, mean annual rainfall measured at the crowns of the test piles was 113 mm. Table 3-2 documents the natural rainfall recorded at the crowns of the UTPs from 2007 through 2011. Several applied rainfall/tracer events were conducted at the crown of the Type III test pile in 2006 and 2007  64  and at the Type III AZLs in 2007 and 2008. These applied events are documented in Tables 3-3 and 3-4. Total annual rainfall at each test pile and AZL is documented in Table 3-5. Total annual rainfall at the test piles varied substantially over the 5 year period; the largest rainfall year was 2008 with 154 mm, and the smallest rainfall year was 2009 with 74 mm. Using the average rainfall of 164 mm, determined by Golder Associates (2008), an average rainfall year was never seen at the test piles. Another important rainfall characteristic that varied substantially year-over-year was the timing and magnitude of the individual rainfall events. Figure 3-6 shows daily rainfall in 2010 and 2011 and is used here as an illustration (the other years are discussed in Section 3.3.2). The majority of the rainfall events in 2010 were small, with the average daily event being 0.5 mm/d (excluding days with no rainfall). Very few events were greater than 5 mm/d and only two larger events in June and July were greater than 10 mm/d. The average daily rainfall event in 2011 was larger than the average event in 2010, at 0.9 mm/d, with many larger (>5 mm) rainfall events occurring later in the rain-season (i.e. August and September). Table 3-6 documents the mean magnitude of daily rainfall events (excluding days with no rainfall) and the number of rainfall days, from 2007 through 2011. Only natural rainfall events are included in Table 3-6. The difference in the timing and magnitude of rainfall events has important implications for evaporation and net infiltration, especially when considering the dramatic changes in both temperature and incoming solar radiation throughout the rain-season at DDMI.  65  3.3.2  3.3.2.1  Net infiltration  Active-zone lysimeter water balance Due to the time taken for the individual AZLs to wet-up, there is no net infiltration  estimate for 2007 and no Type I East AZL estimate in 2008. In 2009, approximately 5 mm of outflow was recorded at both the Type I West and Type III East AZLs prior to significant rainfall occurring. This volume represents remobilization of stored water near the drains and has been subtracted from the 2009 water balances and added to the 2008 water balances. There is also no Type III West AZL water balance estimate in 2011, due to an unknown volume of outflow by-passing the tipping bucket. Table 3-7 summarizes the annual net infiltration calculated at each AZL, while Table 3-8 summarizes the annual percent net infiltration calculated at each AZL. Percent net infiltration refers to the proportion of total annual rainfall that infiltrated in a given year (net infiltration divided by total rainfall). In 2008, percent net infiltration at the Type I West AZL was 57% of natural rainfall, while percent net infiltration at the Type III East and West AZLs was 57% and 54% of rainfall (includes applied events), respectively. In 2009, percent net infiltration was the lowest of any year and had the widest range from 8% at the Type I East AZL to 18% at the Type I West and Type III East AZLs. In 2010, the range in percent net infiltration was from 38% at the Type I East AZL to 43% at the Type III East AZL. In 2011, percent net infiltration was similar to 2008, with a range from 54% at the Type I East AZL to 60% at the Type III East AZL. These results show that percent net infiltration varied significantly from 2008 through 2011 (from 8% in 2009 to 60% in 2011). This variation means that interpreting  66  an average value of percent net infiltration and applying it to multiple years should be done with caution.  3.3.2.2  FAO-56 Penman-Monteith method FAO-PM calculated net infiltration for 2008 through 2011 are compared with AZL  water balance estimates of net infiltration in order to assess the applicability of the FAO-PM method to the AZLs. Figure 3-7 is an example of the output from the FAO-PM method; daily rainfall, cumulative rainfall, reference evaporation, actual evaporation, and net infiltration are plotted for the Type I AZLs in 2008. Graphical FAO-PM output for the remaining years is documented in Appendix A. Table 3-9 compares FAO-PM calculated net infiltration and percent net infiltration with the average AZL water balance values for 2008 through 2011. Figure 3-8 graphically compares the FAO-PM calculated net infiltration with the minimum, maximum, and average AZL water balance estimates. The FAO-PM calculation reasonably predicts net infiltration at the AZLs from 2008 through 2011, using calibrated values for REW (1 mm) and Ze (0.05 m) and keeping these parameters constant over those years. The FAO-PM net infiltration estimates are 94%, 105%, 91%, 108%, and 99% of the average AZL water balance estimates for 2008 (Type I AZL), 2008 (Type III AZL), 2009, 2010, and 2011, respectively. The success of the FAO-PM method over multiple years indicates that a simple, indirect reference evapotranspiration computation coupled with a coefficient-based calculation of actual evaporation is able to predict net infiltration into waste-rock in a northern climate. However, the calibration of coefficients to independent estimates of net infiltration, which was possible in this research setting using the AZLs, may not be possible  67  for many mine sites. Section 3.3.3 discusses the sensitivity of the net infiltration estimates to the calibrated coefficients.  3.3.2.3  Factors influencing net infiltration From 2007 through 2011, percent net infiltration varied substantially due to  evaporation and net infiltration being strongly dependent on the timing and magnitude of rainfall events. The dependence of net infiltration on timing of rainfall events is a function of: 1) the antecedent moisture content in the soil evaporation layer, which controls drainage of water past the zone of evaporative influence, 2) the magnitude of ETo which controls the energy available for evaporation, and 3) the state of the waste rock; whether or not the surface is frozen. The magnitude of a rainfall event controls how much and how quickly the water infiltrates past the soil evaporation layer. The FAO-PM method calculates how much water infiltrates past the evaporation layer, but the effects of rainfall-magnitude on flow velocities within the evaporation layer are not taken into account (discussed in Section 3.3.3). Figures 3-9 to 3-13 are plots of FAO-PM calculated net infiltration and AZL outflow (primary y-axis), and daily rainfall (secondary y-axis). It is important to understand that while the FAO-PM method calculates net infiltration for each time-step, the AZL outflow only approximates net infiltration at the end of the flow-season, when change in storage is zero. Therefore, FAO-PM net infiltration calculated for individual rainfall events should not be compared with AZL outflow response to individual rainfall events. Based on the FAO-PM method, small rainfall events (<5 mm/d) did not contribute significantly to net infiltration in June, July, and August when ETo was highest. Instead, net infiltration from these small events was restricted to periods early and late in the rain-season  68  when decreased solar radiation minimized ETo. This is illustrated best using 2009 and 2010 as examples. In 2009 and 2010 (Figures 3-11 and 3-12), net infiltration from rainfall events <5 mm/d occurred late in the rain-season, while small events in June, July, and August were fully evaporated. Based on the FAO-PM method, net infiltration in June, July, and August was primarily restricted to rainfall events greater than 5 mm/d, with more significant contribution coming from rainfall events approaching 10 mm/d. This is illustrated well in 2008 (Figure 3-9), where the 14 mm, 13 mm, and 33 mm rainfall events on June 15th, August 13th, and August 24th, respectively, were the main contributors to net infiltration during periods of high ETo. This is also illustrated well in 2009, where the FAO-PM estimate suggests net infiltration occurred in the summer from the three >5 mm/day rainfall events in July. 2008 and 2011 (Figure 3-9 and 3-11) are good examples of years where closely spaced, large rainfall events kept antecedent moisture contents near field capacity in the evaporation layer and facilitated net infiltration. In addition, 2008 and 2011 also illustrate the significant contribution that large, late season rainfall events can have to annual net infiltration in a northern climate. There are two distinct annual AZL outflow responses based on annual rainfall pattern. In years with small rainfall events (<5 mm/d) in June, July, and August, AZL outflow response to these individual rainfall events is delayed in comparison to the FAO-PM calculation (i.e. 2009 and 2010). Conversely, in years with larger rainfall events (>5 mm/d) in June, July, and August, AZL outflow response to these individual rainfall events closely follows the FAO-PM calculation (i.e. 2008 and 2011). The proposed explanation for this is a topic in Chapter 4, and relates to pressure waves from larger rainfall events displacing resident porewater in the matrix.  69  3.3.2.4  Application of net infiltration estimates to the crowns of the uncovered test  piles No multi-year, direct estimate of net infiltration was possible at the uncovered test piles to assess the applicability of either the AZL water balance estimate of net infiltration or the FAO-PM method. Instead, near surface volumetric moisture content response beneath the crowns of the UTPs is compared to outflow response at the AZLs to provide a qualitative assessment of net infiltration. Volumetric moisture contents in Figures 3-14 to 3-17 are plotted for periods when the matrix-material surrounding the TDR sensors was thawed. There is an excellent match between the timing of AZL outflow and TDR sensor response at the uncovered test piles. Increases in volumetric moisture content beneath the crowns of the UTPs corresponded with outflow at the AZLs, while downward trends in moisture content (representing net downward drainage) corresponded with periods of no/minimal AZL outflow. This is an important observation since the depths of the TDR sensors (1 m and 2 m) bound the depths of the AZL drains (1.45 m to 1.7 m). These observations cannot be used to comment on the volumes of net infiltration into the crowns of the UTPs; however, they do show that net infiltration into the crowns of the UTPs occurred concurrently to net infiltration into the AZLs. In addition to near-surface TDR sensors being used to assess applicability of the AZL estimates to the crowns of the UTPs, fluxes past tensiometer profiles at the crown of the Type III test pile were used to assess the applicability of the FAO-PM method to the Type III test pile in 2007. Figure 3-18 compares FAO-PM calculated net infiltration and Type III test pile tensiometer flux in 2007. Tensiometer data collection began on August 2nd. As a result, 30 mm of flux (the FAO-PM calculation as of August 2nd) was used as the pre-measurement  70  flux for graphical comparison to the FAO-PM calculation after August 2nd. The termination of the flux data corresponds with tensiometer removal on September 15th. There is a reasonable match between the FAO-PM calculated net infiltration and tensiometer flux between August 2nd and September 15th, which supports the use of the FAO-PM method in predicting net infiltration into the crown of the Type III test pile. Note that the observed response at the tensiometers is delayed in comparison to the PAO-PM calculation, due to the 0.6 to 1.2 m flow path of the infiltrating water to the tensiometers, and that the tensiometers were removed prior to the entire net infiltration event from the September 13th, 29 mm rain event concluding. Due to the reasonable performance of the FAO-PM method in the prediction of net infiltration at the AZLs and net infiltration at the crown of the Type III test pile during tensiometer deployment, the FAO-PM method estimates are considered reasonably applicable, as a first-order estimate, to the crowns of the UTPs. Similar surface textures of the upper surfaces of the AZLs and crowns of the UTPs, the match in timing of near-surface moisture content response beneath the crowns of the UTPs and AZL outflows, and the use of identical Ke parameters for both the AZL and the UTP FAO-PM calculations further support the applicability of the FAO-PM method to the crowns of the UTPs. Table 3-10 summarizes the first-order estimates of net infiltration at the crowns of the UTPs from 2006 through 2011. The 2006 and 2007 values are FAO-PM calculated results. The 2006 value is only for the applied rainfall events at the crown of the Type III test pile. The 2008 through 2011 values correspond to the average net infiltration based on AZL water balances.  71  3.3.3  Uncertainty, limitations, and sensitivity The accuracy of net infiltration estimates based on AZL water balances is affected by  uncertainty in AZL outflows. Uncertainty in AZL outflows is dominated by the uncertainty in the volume of water removed for geochemical sampling. The sampled water was removed before it could pass through a tipping bucket (to preserve the chemistry of the water), and thus, the sampled volume had to be manually added back into the dataset. Typically, 0.5 L of effluent was removed per sampling period. However, uncertainty in sampling volumes arises due to differences in individual sampling practices, as well as a lack of rigorous documentation during low-flow periods when a full 0.5 L was not sampled. A typical lowest volume sampled during low-flow periods was 0.25 L at the AZLs. In 2010, an estimate of the sampling days during low flow periods multiplied by 0.25 L and 0.5 L yielded 3 L and 6 L of estimated outflow, respectively. This corresponds to approximately 3.5 % and 7 % of total outflow in 2010, respectively, and illustrates a typical annual uncertainty in AZL outflow. The limitations of the FAO-PM method reflect the use of a method developed for soils to heterogeneous waste-rock. The effects of rainfall magnitude on preferential flow mechanisms are not taken into account in the FAO-PM method. For example, in the FAOPM method, all water that exceeds field capacity within the evaporation layer is calculated to contribute to deep percolation. In heterogeneous waste-rock; however, rainfall with intensities greater than the saturated hydraulic conductivity of the matrix may result in ponding and the subsequent activation of preferential pathways that result in the rapid movement of water past the zone of evaporative influence, at volumes potentially greater than under matrix flow conditions. Chapter 4 discusses preferential flow at the AZLs and uncovered test piles.  72  Parameter sensitivity was carried out to determine the sensitivity of the FAO-PM method to uncertainty in the depth of the soil evaporation layer (Ze) and the maximum depth of water that can be evaporated from the evaporation layer without restriction (REW). The sensitivity of the FAO-PM method to changes in these parameters was completed by holding one parameter constant and varying the other by a factor of 2 above and below its best-fit value. Table 3-11 summarizes the sensitivity analysis and Figure 3-19 graphically compares net infiltration calculated using the different parameter values. The FAO-PM method is most sensitive to changes in Ze. Decreasing Ze by a factor of 2 increased net infiltration by 25 %, while increasing Ze by a factor of 2 decreased net infiltration by 18 %. Changing Ze affects the calculation of TEW, the maximum depth of water that can be evaporated from the completely wetted evaporation layer. By changing Ze, TEW changes according to Equation 3-7, (  )  3-7  where‎θFC and‎θWP are field capacity and wilting point, respectively. The FAO-PM method was much less sensitive to changes in REW. Decreasing REW by a factor of 2 increased net infiltration by 2 %, while increasing REW by a factor of 2 decreased net infiltration by 5%. The sensitivity of the FAO-PM method to Ze suggests that this method may not be ideal for predicting net infiltration into waste-rock when no direct means of comparison and calibration are available. Ze is a difficult parameter to measure; Allen et al. (1998) recommend that a Ze of 0.1 m to 0.15 m be used for most soils. A Ze of 0.15 m corresponds to an increase of a factor of 3 above the best-fit value, and underestimates net infiltration by 28 % in 2008. Assuming 0.15 m is the maximum Ze that a user of the FAO-PM method  73  would select, this study suggests that the FAO-PM method calculation may be accurate to within ±28 % for waste-rock of similar properties in a northern climate.  3.4  Conclusions The following conclusions regarding net infiltration and the FAO-PM method are  highlighted:  1. There was significant variability in annual percent net infiltration at the active-zone lysimeters over a five year period. Percent net infiltration was as low as 8% of total rainfall in 2009 and as high as 60% of total rainfall in 2011. Therefore, caution should be used when applying an average value for annual percent net infiltration.  2. Parameter calibration of the FAO-PM method to direct measurement of net infiltration, based on active-zone lysimeter water balances, resulted in an indirect computation that reasonably predicted net infiltration into the active-zone lysimeters for multiple years without changes to parameters between years. The least accurate FAO-PM net infiltration estimate (i.e. 2009) was 91% of the average AZL water balance estimate. However, the sensitivity of the FAO-PM method to the depth of the evaporation layer (Ze) means that this method should be used with caution when assigning recommended (e.g. Allen et al., 1998) or literature values of Ze to wasterock. Using the maximum suggested Ze (0.15 m), by Allen et al. (1998), the FAO-PM method was only accurate to ±28% of the direct estimate of net infiltration at the active-zone lysimeters.  74  3. Net infiltration into the upper surface of uncovered waste-rock in the Canadian Arctic is largely dependent on the timing and magnitude of rainfall events. The following are observations from the FAO-PM method estimates at the active-zone lysimeters: i.  Net infiltration of small rainfall events (typically <5 mm/day) was restricted to periods early (May) and late in the rain-season (September to mid-October) when ETo was minimal.  ii.  Net infiltration in June, July, and August was restricted to larger rainfall events (typically >5 mm/day).  4. Estimates of net infiltration from AZL water balances and from the FAO-PM method are considered applicable to the crowns of the uncovered test piles as first-order estimates.  75  3.5  Figures  Figure ‎3-1  Type I test pile crown, facing north. (a): Snow covered crown on March 21, 2012.  (b): Crown with minimal snow on May 2, 2012, due to wind-scouring and sublimation prior to thawing conditions.  76  Figure ‎3-2  Type III test pile crown, facing west. (a): Snow covered crown on March 21, 2012.  (b): Crown with minimal snow on May 2, 2012, due to wind-scouring and sublimation prior to thawing conditions.  77  Figure ‎3-3  Surface texture at the upper surface of the AZLs (a), Type I test pile crown (b), and Type III test pile crown (c). Columns illustrate  similar surface textures of the different locations. The white square in each image is 1 m on each side.  78  Figure ‎3-4  2010 average monthly incoming solar radiation recorded at the DDMI meteorological  station. Source: DDMI Environment Dept.  Figure ‎3-5  2010 air temperature at the DDMI meteorological station  79  Figure ‎3-6  Daily rainfall measured at the crowns of the test piles in 2010 (a) and 2011 (b)  80  Figure ‎3-7  FAO-PM method output for the Type I active-zone lysimeters in 2008  81  Figure ‎3-8  Comparison of yearly active-zone lysimeter (AZL) net infiltration estimates and FAO-PM calculated net infiltration  82  Figure ‎3-9  FAO-PM calculated net infiltration and Type I active –zone lysimeter (AZL) outflow for  2008. AZL outflow at the end of the year is approximately equal to net infiltration.  Figure ‎3-10  FAO-PM calculated net infiltration and Type III active-zone lysimeter (AZL) outflow  for 2008. AZL outflow at the end of the year is approximately equal to net infiltration.  83  Figure ‎3-11  FAO-PM calculated net infiltration and active-zone lysimeter (AZL) outflow for 2009.  AZL outflow at the end of the year is approximately equal to net infiltration. Note: Different primary yaxis scale than Figures 3-9, 3-10, 3-12, and 3-13.  Figure ‎3-12  FAO-PM calculated net infiltration and active-zone lysimeter (AZL) outflow for 2010.  AZL outflow at the end of the year is approximately equal to net infiltration.  84  Figure ‎3-13  FAO-PM calculated net infiltration and active-zone lysimeter (AZL) outflow for 2011.  AZL outflow at the end of the year is approximately equal to net infiltration.  85  Figure ‎3-14  Type I West active-zone lysimeter (AZL) outflow and near surface volumetric moisture  content response at the uncovered test pile crowns for 2008.  Figure ‎3-15  Active-zone lysimeter (AZL) outflows and near surface volumetric moisture content  response at the uncovered test pile crowns for 2009  86  Figure ‎3-16  Active-zone lysimeter (AZL) outflows and near surface volumetric moisture content  response at the uncovered test pile crowns for 2010  Figure ‎3-17  Active-zone lysimeter (AZL) outflows and near surface volumetric moisture content  response at the uncovered test pile crowns for 2011  87  Figure ‎3-18  FAO-PM calculated net infiltration and flux past tensiometer profiles at the crown of  the Type III test pile for 2007. Truncation of tensiometer data represents removal of instrumentation.  Figure ‎3-19  FAO-PM method sensitivity analysis using the depth of the evaporation layer (Ze) and  the maximum depth of water that can be evaporated from the evaporation layer without restriction (REW). The best-fit net infiltration corresponds to the best-fit Ze and REW values for all years.  88  3.6  Tables  Table ‎3-1  Key parameters used in calculating Ke  Parameter Field Capacity (m3/m3) Wilting Point (m3/m3) Ze (m) TEW (mm) REW (mm)  Table ‎3-2  Natural rainfall measured at the crowns of the test piles  Year 2007 2008 2009 2010 2011 Table ‎3-3  Value 0.1 0.06 0.05 3.5 1  Natural rainfall (mm) 92.4 154.4 74.0 97.5 145.5  Applied rainfall at the crown of the Type III test pile (area of influence = 20mx30m)  Applied rainfall at Rainfall intensity Date Type III test pile crown Tracer (mm/hr) (mm) 20/Sep/2006 24.2 10.2 24/Sep/2006 19.0 9.5 26/Sep/2006 14.7 9.8 17/Aug/2007 16.1 7.6 D2O 4/Sep/2007 15.0 8.6 13/Sep/2007 29.2 7.7 Cl & Br-  89  Table ‎3-4  Applied rainfall at the Type III active-zone lysimeters (AZLs)  Applied rainfall at Rainfall intensity Date Type III AZLs Tracer (mm/hr) (mm) 4/Aug/2007 23.1 8.2 5/Aug/2007 27.7 12.3 6/Aug/2007 14.1 8.2 1/Sep/2007 14.8 8.4 19/Sep/2007 11.6 10.8 6/Aug/2008 10.6 15.2 12/Aug/2008 14.9 16.3 Cl-  Table ‎3-5  Table ‎3-6  Total annual rainfall at the test piles and active-zone lysimeters (AZLs)  Year  Type III test pile crown (mm)  Type III AZLs (mm)  2007 2008 2009 2010 2011  152.7 154.4 74.0 97.5 145.5  183.7 179.9 74.0 97.5 145.5  Type I test pile Type III test pile batters Type I AZLs (mm) 92.4 154.4 74.0 97.5 145.5  Rainfall (natural) statistics  Date All Years 2007 2008 2009 2010 2011  Mean magnitude of daily rainfall events (mm) 0.3 0.5 1.0 0.5 0.5 0.9  Number of rainfall days 294 50 63 65 60 56  90  Table ‎3-7  Year 2008 2009 2010 2011  Table ‎3-8  Net infiltration (NI) calculated based on active-zone lysimeter water balances  Rainfall (mm) 154 (TI) 180 (TIII) 74 98 146  Type I East NI (mm)  Type I West NI (mm)  Type III East NI (mm)  Type III West NI (mm)  -  88  102  97  6 37 79  13 41 86  13 42 88  11 40 -  Percent net infiltration (PNI) calculated based on active-zone lysimeter water balances  Rainfall (mm) 154 (TI) 2008 180 (TIII) 2009 74  Year  Type I East Type I West PNI PNI 57%  57%  54%  8%  18%  15%  43%  41%  60%  -  2010  98  38%  2011  146  54%  59%  Year 2008 2009 2010 2011  Type III West PNI  -  18% 42%  Table ‎3-9  Type III East PNI  Comparison of AZL and FAO-PM net infiltration (NI) and percent net infiltration (PNI)  Rainfall (mm) Type I: 154 Type III: 180 74 98 146  Average AZL NI (mm) 88 100 11 40 84  Average AZL PNI 57% 56% 15% 41% 58%  FAO-PM NI (mm) 83 105 10 43 83  FAO-PM PNI 54% 58% 14% 44% 57%  91  Table ‎3-10  First-order estimates of net infiltration (NI) and percent net infiltration (PNI) into the  crowns of the uncovered test piles  Year 2006 (Type III) 2007 (Type I) 2007 (Type III) 2008 2009 2010 2011  Table ‎3-11  Rainfall (mm) 58 (applied events) 92 153 154 74 98 146  NI (mm)  PNI  Method  51  n/a  FAO-PM  40 92 88 11 40 84  43% 60% 57% 15% 41% 58%  FAO-PM FAO-PM AZL water balance (average) AZL water balance (average) AZL water balance (average) AZL water balance (average)  FAO-PM method sensitivity analysis (using 2008 natural rainfall)  Ze Ze Ze  Best fit (to all years) Decrease by factor of 2 Increased by factor of 2  Parameter value 0.05 0.025 0.1  REW REW REW  Best fit (to all years) Decrease by factor of 2 Increased by factor of 2  1 0.05 2  Parameter  Net infiltration (mm) 83 103 68 83 85 79  92  Chapter 4: Wet-up and fluid flow in waste-rock undergoing annual freezethaw  4.1  Introduction An understanding of the processes governing fluid flow through waste-rock can be  essential to predicting solute loadings to the environment at mine sites. Quantifying the hydraulic properties of the waste-rock material is a central challenge in understanding and characterizing fluid flow through waste-rock (Smith and Beckie, 2003). Neuner et al. (2012) focused on quantifying the hydraulic properties of Type I and III waste-rock at DDMI (summarized in section 2.4). The focus of this chapter is to describe fluid flow within and outflow from the uncovered test piles. Section 1.4.2 outlined fluid flow processes through waste-rock, including the distinction between different fluid flow mechanisms (matrix flow and preferential flow). An additional concept critical to understanding the hydrologic behaviour of the uncovered test piles is the concept of a wetting front. A wetting front is defined here as a moisture front that travels downwards through the unsaturated zone following an infiltration event. In the test piles, wetting fronts are detected as increases in moisture content by time domain reflectrometry (TDR) sensors. Infiltrating water can also generate a pressure wave that propagates downwards through porous media at velocities much greater than the velocity of a particle of water (Rasmussen et al., 2000; Neuner et al., 2012). The velocity of a particle of water is equivalent to the average velocity of a conservative tracer (tracer velocity). Under applied rainfall conditions, both Rasmussen et al. (2000) and Neuner et al. (2012) observed pressure wave velocities that were up to 1000 times faster than tracer velocities in unsaturated saprolite and waste-rock, respectively. Propagating 93  pressure waves may displace resident pore-water in the waste-rock, resulting in an apparent wetting front (i.e. measured by TDR sensors as increases in moisture content) that travels faster than tracer velocities (Nichol et al., 2005; Neuner et al., 2012). The objectives of this chapter are to: 1) use data from moisture content sensors to describe the freeze-thaw cycle, initial matrix wet-up, and wetting front movement in the uncovered test piles, and 2) use outflow data to describe the evolution of outflow over annual and multi-year time-scales, as well as to estimate crown and batter contributions to outflow.  4.2  Methods A description of the study site was presented in Chapter 2; Sections 2.1 and 2.2  provided an overview of DDMI and the Test Piles Research Area. The data presented in this chapter pertain to the uncovered test piles (Type I and Type III) and active-zone lysimeters (AZLs).  4.2.1  Test piles Construction and instrumentation of the uncovered test piles was described in  Sections 2.2 and 2.3. The following points are highlighted:   The Type I and Type III uncovered test piles have similar dimensions: 15 m height, 50 m by 60 m (3000 m2) basal areas, and approximately 1350 m2 crowns.    The surface area of the batters is larger than the surface area of the crown for each uncovered test pile: 55% of the basal area underlies the batters, while 45% of the basal area underlies the crown.  94    The matrix-fraction (<5 mm fraction) of the uncovered test piles constitutes approximately 18% of the bulk waste-rock by volume. The average porosity of the matrix material and the average bulk-porosity of the waste-rock piles are similar (~0.25).    Average grain-size of the cobble and boulder fraction increases towards the base of the test piles, due to gravitational segregation during construction.    All water reaching the base of each test pile is intercepted by a water collection system, consisting of an HDPE liner and perforated drain pipe, and routed to tipping buckets for measurements of flow and geochemistry.  The following points are highlighted regarding the water collection systems at the uncovered test piles:   Six 4 m by 4 m and two 2 m by 2 m basal collection lysimeters (BCLs) underlie the crown of each test pile and two 2m by 2m BCLs underlie the batters at each test pile (Figures 2-6 and 2-7, Chapter 2). At each uncovered test pile, the BCLs comprise 4 % of the total basal area and 8 % of the basal area that underlies the crown.    All drain lines are heat-traced to ensure that water remains liquid as it moves out of the test piles. In addition, the base of each BCL is heat-traced.    The drainage schemes at the Type I and Type III uncovered test piles differ. At the Type III test pile, water is directed across the basal liner towards drain lines at the outer edges of the pile, while at the Type I test pile water is directed along the basal liner towards a drain line that runs through the centre of the pile. Additionally, the 0.3 m of crush material protecting the liner and drain line at the base of the Type I test  95  pile has significantly more fines than the crush material at the base of the Type III test pile. Section 2.3.4.2 described the TDR sensors installed in the test piles. Figures 4-1 and 4-2 show locations of TDR sensors within the Type I and III test piles, respectively. The following points are highlighted:   All TDR sensors were installed beneath the crowns of the uncovered test piles, meaning that moisture content data is not available beneath the batters.    The TDR sensors were placed in hydraulic contact with fine-grained (2.5 cm minus) material. Therefore, water movement through the larger-grained fraction is not directly monitored.    The dielectric permittivity of ice and liquid water are different, which enables the freeze-thaw‎cycle‎of‎the‎test‎piles‎to‎be‎interpreted‎by‎observing‎the‎‘apparent‎ volumetric‎moisture‎content’‎(VMCa) response about the phase change of water. Here, VMCa refers to the apparent moisture content measured at the TDR sensors regardless of thawed or frozen conditions.    The uncertainty in volumetric moisture content, resulting from the application of the calibration curve to sensors surrounded by large rocks and/or air-filled pores, may result in an underestimation of moisture content by as much as 0.07 m3/m3 (Neuner, 2009). Thirteen TDR sensors survived construction of the Type III test pile, while only four  sensors survived in the Type I test pile. The TDR sensors take measurements every 30 minutes. In order to process the data for graphing purposes, a moving average filter was applied to the data using a 12 data point (6 hour) averaging function. TDR sensors were also 96  used to observe wetting front propagation and to calculate wetting front velocities. Using similar methods to Neuner et al. (2012), wetting front velocities were calculated as the depth to the TDR sensor divided by the average time of wetting front arrival at that sensor. The average time of wetting front arrival was calculated by taking an average of 1) the time from rainfall to initial increase in moisture content and 2) the time from rainfall to the peak in moisture content.  4.2.2  Active-zone lysimeters Section 2.2.2 described the construction and instrumentation of the active-zone  lysimeters (AZLs). The following points are highlighted:   Three ECH2O sensors, installed adjacent to the AZLs at 0.3 m, 0.6 m, and 0.9 m depths, are used to measure temperature and moisture content of the matrix.    The Type I and Type III AZLs have different dimensions. Type I AZLs have surface collection areas of 3.6 m2 and zero-tension drains at 1.45 m depth. The Type III AZLs have surface collection areas of 2.1 m2 and zero-tension drains at 1.7 m depth.  97  4.3  Results and Discussion  4.3.1  Active-zone formation Active-zone formation is an issue unique to waste-rock stockpiles in cold climates.  From a hydrological standpoint, the formation of the active-zone is important to characterize because it controls the regions of a stockpile that are hydrologically active.  4.3.1.1  Active-zone lysimeters Figure 4-3 presents ECH2O temperature data at the AZLs from 2007 through 2011.  The freeze-thaw response at the AZLs followed the annual trend in ambient air temperature. At the onset of above-freezing air temperatures, a thaw front starts to propagate through the waste-rock reaching the 0.3 m, 0.6 m, and 0.9 m depth sensors in order. Similarly, at the onset of below-freezing air temperatures, a freeze front starts to propagate through the wasterock reaching the 0.3 m, 0.6 m, and 0.9 m depth sensors in order. The 0.9 m sensor is the deepest sensor at the AZLs and is used to indicate when over half the AZL depth is frozen or thawed. Table 4-1 documents the dates of freeze and thaw at the 0.3 and 0.9 m sensors. From 2007 through 2011 the 0.9 m depth sensor recorded the onset of thawed conditions between May 27th and June 10th. For the same time period (2007 through 2011), the 0.9 m depth sensor recorded the onset of frozen conditions between October 29th and November 4th. On average, the AZLs were thawed to a depth of 0.9 m for 152 days a year. From 2007 through 2011, freeze- and thaw fronts propagated at an average velocity of 0.06 m/d and 0.05 m/d, respectively, from the 0.3 m to the 0.9 m sensor. Outflow from the AZLs was recorded in all  98  years following initial matrix wet-up, indicating that the active-zone developed to the full depth of the AZLs during the summer months.  4.3.1.2  Test piles The development of the active-zone was also monitored at the uncovered test piles.  Thermistor strings located within the uncovered test piles are used to monitor internal pile temperatures (Pham, 2012). In addition, the freeze-thaw response at the uncovered test piles can be described using TDR sensor response. Figure 4-4 compares VMCa at the 31S2 TDR string with thermistors installed on the same tip face as the TDR sensors and at corresponding depths (31S5 thermistor string). Despite the TDR sensors being located 3 m closer to the centre of the test pile than the thermistor string, there is a clear correlation between the onset of above freezing conditions at the thermistors and the rapid increase in VMCa measured by the TDR sensors. Similarly, there is a clear correlation between the onset of below freezing conditions at the thermistors and the rapid decrease in VMCa measured by the TDR sensors. This behaviour demonstrates that the TDR sensors can be used to interpret and monitor the depth of the active-zone and the timing of its response. Data from the test piles indicates that the TDR sensors are less reliable at monitoring freeze-thaw when the waste-rock surrounding them is dry. This is illustrated well at the Type I test pile from 2009 through 2010 prior to the matrix surrounding the 9 m sensor being wetted-up (Figure 4-5). It is difficult to determine, from the TDR data alone, that the 9 m depth location thawed during the 2009 and 2010 summers even though thermistor data showed that it did. This is because the dielectric permittivities of ice and air are very similar, resulting in an insignificant change  99  in VMCa between below- and above-freezing conditions when pore space is predominantly air-filled. Figure 4-6 presents VMCa data from the 31S2 TDR string in the Type III test pile for 2009 (the matrix material surrounding these sensors had wet-up prior to 2009). VMCa; at 1 m, 3 m, 5 m, and 9 m depths; is plotted on the primary y-axis (in %), while daily rainfall is plotted on the secondary y-axis (in mm). From January to mid-June, frozen pile conditions resulted in low VMCa measurements between 2.5-7.5 % (recall, true VMC is not measured due to an ice-phase being present). Beginning in mid-June, a rapid rise in VMCa was observed starting at the 1 m depth sensor and followed in order by the 3 m, 5 m, and 9 m depth sensors. This corresponded to a thaw front moving inwards through the pile and follows the same trend with depth seen at the AZLs. Beginning in November, VMCa decreased rapidly as the test pile underwent freezing; however, the freezing trend observed at the test piles differed from the trend observed at the AZLs. At the AZLs, the waste-rock froze from the surface downwards, whereas at the uncovered test piles the deep, central locations (i.e. 9 m depth sensor; Figure 4-4) froze prior to the locations nearer the surface (i.e. 3 m – 5 m depth sensors). There are two factors that are responsible for this trend: 1) air convection at the highly permeable toes of the piles and 2) permafrost under the basal liner (Pham, in communication). Air convection at the highly permeability toes is the dominant factor in this case. The batters of the test piles are located nearest to the atmosphere and, consequently, they freeze and thaw prior to the more central regions of the piles. The sloping surface of the batters results in the thinner toes of the test piles freezing and thawing to the liner prior to the thicker portions of the batters.  100  An absence of thermistors between 11 m depth and the base of the 15 m uncovered test piles creates uncertainty as to whether the active-zone develops into the full extent of the uncovered test piles each year. Thermistors at 11 m depth indicate that this depth of the uncovered test piles remained thawed for approximately 67 days at Face 1 and 46 days at Face 4, on average, each year (Pham, in communication). Outflow from the Type III BCLs indicates that the waste-rock surrounding the BCLs thawed briefly for up to 2 months a year, from September to November, and shows that at least a portion of the basal liner underlying the crowns thawed each year. Thermal modeling by Pham (2012) suggests that some portion of the base of the test piles may remain frozen year round.  4.3.2  Initial matrix wet-up The initial wet-up of the matrix material at the AZLs and uncovered test piles was  recorded by ECH2O and TDR sensors and by the initiation of outflow. Only the matrix wetup below the crowns, and not within the batters of the uncovered test piles, was recorded due to TDR sensor placement. The matrix fraction of the run-of-mine waste-rock was emplaced below residual saturation during the construction of the AZLs and uncovered test piles (Neuner et al., 2012), giving it a measurable capacity to store water prior to further downward migration of water.  4.3.2.1  Active-zone lysimeters Matrix wet-up of the Type III AZLs is discussed by Neuner et al. (2012), while the  Type I AZLs are discussed here. The ECH2O sensors and tensiometers adjacent to the AZLs could not be used to interpret fluid flow at the Type I AZLs in 2007 and 2008, due to applied  101  rainfall at the Type III AZLs (the applied rainfall footprint covered the region of influence over the sensors). Outflow was first observed in late June at the Type I West AZL and late August at the Type I East AZL. At the initiation of outflow, net infiltration at the Type I West AZL was 51 mm, while net infiltration at the Type I East AZL was 107 mm (net infiltration estimate derived from analysis in Chapter 3). The following estimate of field capacity is discussed in terms of bulk waste-rock for comparison to estimates by Neuner et al. (2012). Field capacity of the Type I AZL bulk waste-rock was estimated to first order using the volume of net infiltration required to initiate outflow and taking into account the initial moisture content of the waste-rock. The initial moisture content of the bulk waste-rock (0.01 m3/m3) was estimated by Neuner et al. (2012) as the initial moisture content of the <40 mm fraction (0.025 m3/m3) multiplied by the fraction of the waste-rock finer than 40 mm (0.35). The calculated bulk field capacity of the Type I AZLs waste-rock was approximately 0.05 to 0.08 m3/m3, which is similar to the bulk field capacity of 0.06 m3/m3 calculated by Neuner et al. (2012) for the Type III AZLs. The bulk field capacity calculation assumes that all water that infiltrated into the AZLs prior to the initiation of outflow acted to wet-up the matrix fraction and did not by-pass the matrix by preferential mechanisms such as macropore flow.  4.3.2.2  Type III test pile The crowns of the Type I and Type III test piles received different total rainfall in  both 2006 and 2007 (Table 3-5, Chapter 3), providing an opportunity to examine matrix wetup under two different rainfall conditions. Collection of TDR data began in September of 2006 at the Type III test pile. Figures 4-7 and 4-8 present VMCa from TDR sensors within  102  the Type III test pile along strings 31S2 and 33N2 respectively (a complete TDR dataset is documented in Appendix A). The evolution of the VMCa from 2006 through 2008 illustrates the initial wet-up phase of the matrix material. Natural and applied rainfall events in 2006 and 2007 resulted in a wetting front reaching a depth of at least 7 m, but not 9 m, before pilefreeze in 2007. At the beginning of August 2008, the 9 m depth TDR sensors recorded the concurrent arrival of a thaw front and wetting front. No TDR sensors were installed between 9 m and the base of the 15 m test pile; however, outflow recorded at multiple BCLs indicates that a wetting front first reached the base of the Type III test pile in September, 2008. Therefore, with the addition of applied rainfall events in 2006 and 2007, it took approximately 25 months (including the periods when the pile was frozen) to wet-up the matrix material from the crown to the base of the 15 m uncovered Type III test pile. Same-depth TDR sensors provide insight into the spatial and temporal variability in wet-up of the matrix material. Figures 4-9, 4-10, and 4-11 compare TDR sensors at 1 m, 7 m, and 9 m depths respectively. These figures show that while all TDR locations recorded wetup (i.e. no locations remained dry), there were differences in the timing of wetting-front arrival at same-depth sensors. The greatest temporal difference in same-depth wetting-front arrival occurred during the matrix wet-up phase and is illustrated in Figures 4-12 and 4-13. A wetting front arrived at 1 m depth in 2006 at the 31N2 and 33N2 sensors, but did not arrive until thaw in 2007 at the 31S2 sensor. Similarly, a wetting front arrived at 7 m depth at least 1 month earlier at the 31N2 location than the 33N2 location in 2007. The VMCa response in the post wet-up period was observed from 2008 through 2011 at the Type III test pile. During this period, the test pile was in a state of dynamic equilibrium, where the moisture content of the matrix surrounding the TDR sensors  103  fluctuated between 15 % and 25 % (0.15 m3/m3 and 0.25 m3/m3) during thawed conditions. Using the average porosity of the matrix material (0.25) suggests the matrix within the Type III test pile had saturation values that fluctuated between 0.6 and 1 during this period.  4.3.2.3  Type I test pile Collection of TDR data began in September of 2007 at the Type I test pile. Figure 4-  14 presents VMCa from all surviving TDR sensors within the Type I test pile along strings 11W2 and 12W2. The 1 m and 3 m sensors did not record significant wet-up until 2008. A wetting front reached at least 6 m depth in November of 2008; however, an additional 3 years elapsed before a wetting front reached 9 m depth in October of 2011. By the end of 2011, the matrix material between 9 m and 15 m depth had not fully wet-up. This was determined by observing BCL outflow response. Outflow from the Type I BCLs did occur in 2008 through 2011; however, it was minimal and coincided with periods when the majority of the pile below the crown was still frozen; BCL outflow was <0.1%, 2%, <0.1%, and <0.1% of total outflow in 2008, 2009, 2010, and 2011 respectively. This may indicate a small amount of preferential flow through ice-free voids, but it is not indicative of wet-up or flow through the matrix. Under natural rainfall conditions, it took approximately 62 months for the Type I test pile to wet-up to a depth of 9 m, which is significantly longer than the wet-up time of the Type III test pile under applied rainfall conditions. Section 4.3.4 continues the discussion of matrix wet-up, after fluid flow has been presented.  104  4.3.3  Fluid flow Fluid flow under the following two different degrees of matrix saturation was  observed: 1. Matrix moisture content below field capacity. 2. Matrix moisture content above field capacity. Neuner et al. (2012) described fluid flow under these two degrees of matrix saturation using data collected during 2006 and 2007. Under the first condition, they concluded that the lack of moisture in the initially dry matrix meant that fluid flow by displacement of porewater was insignificant and that the wetting front velocity of initial infiltration events was approximately equal to fluid velocity. No evidence of significant preferential flow was observed at either the AZLs or uncovered test piles throughout the initially dry 2006 period, even during the high intensity applied rainfall events conducted at the Type III test pile. Under the second condition, Neuner et al. (2012) concluded that wetting fronts observed by TDR sensors reflected the influence of pressure waves that displaced resident porewater in the matrix. This mechanism was observed at both the AZLs and the uncovered test piles. During applied rainfall events at the Type III AZLs in 2007, Neuner et al. (2012) observed that the specific conductance of the outflow water remained high and that wetting front velocities were approximately 1000 times greater than the water flux. This indicated that the wetting front velocity reflected a propagating pressure wave instead of fluid velocity and that pore-water displacement in the matrix was the dominant mechanism for the observed increases in outflow and not preferential mechanisms. At the Type III test pile, moderate rainfall events in 2007 did not generate preferential flow; however, evidence of preferential flow (in this case, likely macropore flow) was observed following an intense applied rainfall 105  event on September 13th, 2007 (29mm, 8 mm/h, 35 year recurrence interval). This applied event incorporated a LiCl and LiBr tracer. Initial SWSS sampling indicated a heterogeneous flow response; higher flow velocity at greater depth; and flow velocities on the order of rainfall rate and higher, while tensiometers measured near zero matric tension at shallow depths (0.6 to 1.2 m) and TDR sensors recorded spatially variable wetting front velocities between same-depth sensors. Under conditions observed by Neuner et al. (2012), wetting fronts migrated at rates of 0.2 to 0.4 m/d in response to common rainfall events and up to 5 m/d in response to intense rainfall events. Initial tracer study results indicated that porewater and non-reactive solutes migrated at rates of < 0.01 to 0.03 m/d in response to common rainfall events and up to 0.70 m/d in response to intense rainfall events. TDR sensors were used to study wetting front migration through the matrix from 2008 through 2011. Tracer studies are not covered in this thesis. Calculated wetting front velocities (from detectable wetting fronts) over this time period are within the range observed by Neuner et al. (2012); however, a lower bound of 0.01 m/d was calculated at shallow depth (1 m) for multiple wetting fronts. Wetting front velocities calculated at same-depth TDR sensors were more similar in the post matrix wet-up period than during the initial matrix wetup period. This suggests that matrix porewater displacement from propagating pressure waves, post wet-up, is more spatially uniform than matrix flow in the initially dry period where porewater displacement does not occur. Spatially variable wetting front velocities (i.e. wetting fronts whose velocities varied significantly between same-depth sensors and with depth) occurred following large intensity rainfall events, while more spatially uniform wetting front velocities occurred following less intense rainfall events. This is best illustrated using moderate to large rainfall events (~15 to 30 mm/d) in 2008 and 2011 that resulted in  106  pressure wave propagation, detectable as wetting fronts, all the way to the deepest 9 m TDR sensors. Velocities were calculated at all TDR probes that recorded wetting front arrival. Moderate 7.4 mm/d and 10.5 mm/d rainfall events on August 15th and 16th, respectively, resulted in a detectable wetting front that propagated at 0.1 to 0.3 m/d. Another moderate rainfall event on September 26th, 2011 (16.2 mm/d) resulted in a wetting front that propagated at 0.2 to 0.6 m/d. Faster wetting front velocities following the September 26th event were a result of the increased intensity of the rainfall event and the higher average antecedent moisture content of the matrix. There was no discernible correlation between wetting front velocity and depth into the pile for theses moderate intensity rainfall events. An example of spatially variable wetting front velocities occurred following a 33 mm rainfall event on August 24th, 2008. This event generated a pressure wave, detectable as wetting front velocities that ranged from 0.1 m/d to 2.5 m/d. There was no correlation between wetting front velocity and depth into the pile for this high intensity rainfall event. The observation of large spatial variation in wetting front velocities following this event is similar to Neuner et al.’s‎(2012)‎observation‎following‎a‎similar-intensity, 29 mm rainfall event in 2007. Spatially variable wetting front velocities following the 29 mm rainfall event coincided with spatially variable tracer velocities indicative of macropore flow response. It is, therefore, likely that some macropore flow occurred between 1 m and 9 m depth following the 33 mm rainfall event in 2008. Outflow from BCLs at the base of the test pile, below the crown, initiated following this event. Chemistry of the outflow water indicates that the outflow represented fluid flow through matrix material, and not preferential mechanisms (discussed in Section 4.3.5.2).  107  Spatially uniform wetting front velocities were observed for rainfall intensities up to and including the 16.3 mm/d event in 2011. From 2007 through 2011 there were three rainfall events with intensities greater than 16.3 mm/d: a ~29 mm/d applied rainfall event in 2007, a ~33 mm/d event in 2008 and a ~20 mm/d event in 2010. Spatially variable wetting front velocities followed both the ~29 mm/d and 33 mm/d event, while the wetting front response following the ~20 mm/d event in 2010 was obscured by the concurrent thawing of the matrix material surrounding the TDR sensors. Therefore, current data indicate that the lowest intensity rainfall event capable of generating wetting front velocities with a high degree of spatial variability is somewhere between 16 mm/d and 29 mm/d at the uncovered test piles. The multi-year VMC dataset at the Type III test pile (Figures 4-7 to 4-11) shows two characteristic VMC responses based on the character of late-season rainfall. This is illustrated most clearly at 7 m and 9 m depth (Figures 4-10 and 4-11). Years with no large, late season rain events (2009 and 2010) show a decrease in VMC at depth throughout the season. Conversely, years with large late-season rain events (2008 and 2011) show VMCs that remain high near the end of the season, even at depth. The response in 2008 and 2011 is due to pressure waves that propagate to at least 9 m depth and displace resident porewater in the matrix. The VMC dataset, from 2008 through 2011, shows that these two characteristic responses are reproducible over multiple years.  4.3.4  Type I vs. Type III test pile matrix wet-up Net infiltration estimates given in Chapter 3 and volumetric moisture contents from  TDR sensors are used to compare the initial wet-up response at the Type I and III test piles.  108  Note net infiltration values calculated via FAO-PM method, and not via AZL water balance, are used here. This is due to the need for mid-season net infiltration estimates, which are not possible using an AZL water balance (Chapter 3). At the Type I test pile a wetting front first reached the 6 m depth sensor in 2008 when FAO-PM calculated net infiltration into the crown was approximately 123 mm. At the Type III test pile a wetting front reached the 7 m depth sensor in 2007 after approximately 143 mm of net infiltration. For a greater depth, a wetting front reached the 9 m depth sensor in the Type I test pile in 2011 after approximately 265 mm of net infiltration, but did not make it all the way to 15 m depth by the end of 2011. At the Type III test pile, a wetting front reached 9 m depth at thaw in 2008 after approximately 159 mm of net infiltration. BCL outflow indicated that a wetting front reached the base of the Type III test pile (below the crown) after approximately 210 mm of net infiltration. The above breakdown indicates that a wetting front first reached 9 m depth in the Type III test pile with 106 mm less net infiltration than at the Type I test pile. Using net infiltration to compare wet-up is rudimentary, since it does not take into account mechanisms of fluid flow. Recall the FAO-PM method completes a water balance within the evaporation layer near the surface to determine evaporation and net infiltration. A possible explanation for the above difference is that the high intensity applied rainfall events at the Type III test pile displaced resident porewater (via pressure wave) deeper than the equivalent volume of net infiltration did at the Type I test pile. The arrival of a wetting front at 9 m depth, in the Type III test pile, at thaw indicates that the applied rainfall event conducted at the end of the 2007 rain season resulted in porewater displacement well past 7 m and closer to 9 m depth. The average daily rainfall intensity prior to the Type I test pile wetting up to 9 m was 1.9  109  mm/d; whereas the average daily rainfall intensity prior to the Type III test pile wetting up to 9 m was 3.4 mm/d. Note, average daily rainfall intensities exclude days with no rainfall.  4.3.5  Outflow A complete outflow dataset, which includes AZL, BCL, and basal drain outflows, is  documented in Appendix A.  4.3.5.1  Active-zone lysimeters Figure 4-15 presents cumulative annual outflow at the AZLs from 2007 through  2011. Recall, the Type I and Type III AZLs have different surface collection areas. Therefore, cumulative annual outflow has been normalized by AZL collection area and reported in millimetres for comparative purposes. The Type III AZLs received applied rainfall events in 2007 and 2008 (Table 3-4, Chapter 3), but all AZLs received the same rainfall from 2009 through 2011. Note that a clogged tipping bucket at the Type III West AZL in 2011 resulted in approximately 16 mm (34 L) of outflow being missed. This volume of water was estimated as the average outflow volume from the other three AZLs over the period of data loss. The first occurrence of outflow was recorded at the Type III AZLs in 2007 and at the Type I AZLs in 2008. Outflow initiated at the Type III East AZL one day prior to outflow at the Type III West AZL. The difference in timing of initial outflow was much greater at the Type‎I‎AZL’s;‎outflow‎initiated‎at‎the‎Type‎I‎West‎AZL‎two‎months‎prior‎to‎outflow‎at‎the‎ Type I East AZL. Total outflow volumes in the first year of outflow were more similar at the Type III AZLs than at the Type I AZLs. A possible explanation is that a greater volume of  110  fine-grained matrix exists within the Type I East AZL, resulting in a larger volume of infiltration needed to wet-up the matrix prior to further downward migration of water occurring. This may also explain why outflow volumes are similar (between all AZLs) after initial matrix wet-up. Table 4-2 presents annual outflow volumes at the AZLs from 2007 through 2011, including dates that flow started and stopped each year. From 2007 through 2011, the years when all AZLs received the same rainfall, outflow was greatest at the Type III East AZL, followed in order by the Type I West, Type III West, and Type I East AZLs. The greatest relative difference in outflow volumes (i.e. the difference in outflow volumes as a fraction of the total rainfall) during this period occurred in 2009 when annual rainfall was the smallest of any year. Figure 4-16 compares daily flow rate recorded at the Type I AZLs in 2011. It can be seen that, while the general pattern of outflow is similar between the two AZLs, daily outflow rates varied substantially. This was the case between all AZLs in all years, and suggests that the AZLs are not a representative elemental volume (REV) for fluid flow even though total annual outflow volumes are similar. An REV is the scale at which parameters are independent of domain size, such that the domain can be treated as a continuum. The daily outflow pattern at the AZLs clearly and closely follows daily rainfall events >5 mm/d. This is particularly apparent in 2011 where periods of rainfall were ideally spread out to observe this. Wagner et al. (2006) found a negative correlation between flow rate and sulfate/dissolved metal concentrations of outflow waters during macropore response. Dissolved metal (Ni) concentrations in outflow waters during 2011 (Bailey, in communication) were positively correlated with outflow rate. This is indicative of matrix flow (Wagner et al, 2006) and suggests that the quick outflow response time at the AZLs  111  represents pressure wave propagation that displaces resident porewater in the matrix, as opposed to preferential flow.  4.3.5.2  Test piles Figures 4-17 and 4-18 present annual outflow from the Type I and Type III test piles  from 2007 through 2011. Outflows are from the main basal drains and do not include outflow from BCLs. Daily outflow is presented on the primary axis (in litres), while daily rainfall is presented on the secondary axis (in millimetres). Table 4-3 summarizes each year at the uncovered test piles. The summary includes the dates that outflow initiated and ceased each year, the outflow recorded via tipping bucket, the interpolated outflow, and the total annual outflow (the sum of recorded and interpolated outflows). Interpolated outflow refers to the estimated volume of water lost due to equipment malfunction and geochemical sampling each year and is detailed in Table 4-4. Figures 4-19 and 4-20 present select outflow from BCLs in the Type III test pile, with daily outflow in litres. Figure 4-19 presents outflow from BCLs underlying the batters (3BNBlys2W/2E), while Figure 4-20 presents outflow from a BCL underlying the crown (3BNClys4E). A comparison of basal drain and BCL outflow data with TDR and temperature data can be used to describe the evolution of outflow throughout a year. Figures 4-21 and 4-22 are used here as an illustrative example. Figure 4-21 presents basal drain outflow from the Type III test pile in 2011 and Figure 4-22 presents Type III BCL outflow from 3BNClys2W/2E and VMCa along the 31S2 TDR string in 2011. From May through June, cumulative outflow totaled approximately 3 m3. This outflow was recorded prior to any significant rainfall events and prior to the central region of the pile being thawed from crown to base. Therefore,  112  outflow during this period originated from: 1) snowmelt infiltrating into the batters of the pile and migrating to the basal drain as the batters thawed and 2) remobilized water that was held in storage over the previous winter. The toe of the batters contributed first due to the short flowpaths to the liner and the migration of the thaw-front inwards. Outflow increased in July as the thaw-front migrated inwards and deeper and more of the pile became hydrologically active. Rainfall infiltration contributed to outflow during this period, through porewater displacement from propagating pressure waves and through short flowpaths to the liner at the highly permeable toes of the pile; sharp increases in daily outflow closely followed rainfall events, which illustrates the short response time at the coarse batters near the toes of the pile. TDR and thermistor data show that the central region (below the crown) of the pile thawed to a depth of 9 m by mid-July and 11 m by mid-August (Pham, in communication). BCLs beneath the crown of the pile began flowing from the beginning to mid-September as the base of the pile thawed. At this time, the active-zone in the pile was at its maximum extent. Outflow ceased at the beginning of November, prior to some of the shallower central locations freezing. The abrupt stoppage of outflow in November is a good example of how quickly the base of the test piles can freeze. Variations in the timing and magnitude of outflow over multiple years is a function of the timing of active-zone formation, snow accumulation and melt, timing and magnitude of rainfall events, and antecedent moisture contents and wetting front locations held in storage over the previous winter. A comparison of outflow water chemistry from BCLs in the Type III test pile (Bailey, 2012) with outflow rates suggests that fluid flow below the crown was primarily through the matrix. Sulfate concentrations that were positively correlated with outflow rate during 2008  113  suggest that matrix porewater displacement, as opposed to preferential flow, is the dominant mechanism during years with the highest intensity rainfall events observed at the test piles. Annual outflow from the Type I test pile was very different in magnitude from outflow at the Type III test pile (Figure 4-23, Table 4-3). The question addressed here is whether or not the applied rainfall to the crown of the Type III test pile can fully explain the difference in outflow volumes between the Type I and III test piles. Using net infiltration estimates from Chapter 3, the difference in net infiltration at the crowns of the Type I and III test piles due to applied rainfall events in 2006 and 2007 was approximately 100 mm. The footprint of the applied rainfall events was approximately 20 m x 30 m at the crown of the Type III test pile (Neuner, 2009), yielding an additional 60 m3 of water into the Type III test pile. Discounting 2007 (since an unknown volume of water discharged out of the Type I test pile), outflow at the Type III test pile was, at the very least, 470 m3 greater than the Type I test pile outflow through 2011. This discrepancy cannot be explained fully by the additional 60 m3 of applied rainfall infiltration at the crown of the Type III test pile. The following are factors that may be contributing to the difference in observed outflow: 1. Differences in drainage pipe layout and protective crush material at the base of the uncovered test piles. 2. Differences in snow accumulation, sublimation, melt, and infiltration at the batters of the uncovered test piles. 3. Differences in net infiltration, originating from rainfall, at the batters of the uncovered test piles. 4. Differences in the degree and activation of preferential flowpaths at the uncovered test piles.  114  5. Leaks in the basal liner or main basal drain of the Type I test pile. Section 2.2.1.1 discussed the differences in drainage scheme between the Type I and III test piles. The finer protective crush material of the Type I test pile, in combination with the main drainline being routed through the centre of the pile, likely resulted in storage of water within the crush prior to the initiation of outflow at the drainline. With no wetting front reaching the base of the Type I test pile below the crown, the crush material underlying the crown would need to be wet-up from water reaching the base of the pile below the batters and then flowing along the liner towards the drainpipe. A simple calculation gives an idea of the volume of water potentially stored in the Type I test pile crush material. The crush material is 0.3 m high and covers an area of 3000 m2, yielding a volume of 900 m3. Assuming a field capacity of 0.1 m3/m3 to 0.15 m3/m3, the volume of water stored within the crush material may be 90 m3 to 135 m3 assuming the entire volume was wetted to fieldcapacity. Another consideration regarding storage in the crush material is the development of a year-round frozen core around a portion of the drainline. The water stored in the crush material surrounding the drainline may act to prolong the length of time the region remains frozen by increasing the energy required to melt the volume of ice surrounding the drainline. Unfortunately, the lack of thermistors between 11 m depth and the base of the test piles means that the presence or absence of a year-round frozen core surrounding the drainpipe cannot be observed. As mentioned earlier, thermal modelling by Pham (2012) suggests that some portion of the base of the test piles may remain frozen year round. If a year-round frozen core exists at the Type I test pile, it could partially explain the difference in outflow volume by restricting the region of the pile that can contribute to outflow. Slope aspect may affect the accumulation, sublimation, and melt of snow at the batters of the test piles. Surface 115  characteristics of the test pile crowns and shallow TDR response below the test pile crowns indicate that net infiltration at the test pile crowns is similar; however, net infiltration of rainfall may be dissimilar at the batters of the uncovered test piles. Finally, leaks or tears in the main basal drain or liner may account for some of the difference in outflow. For example, a disconnected drainpipe at the Type I test pile in 2007 resulted in an unknown volume of outflow being missed. The above are possible explanations and further investigation is required to further qualify and quantify these possibilities. Due to the differences in drainage pipe layout, protective crush material, and applied rainfall at the Type I and III test piles, it is not possible to determine whether the test piles are large enough to be an REV for fluid flow.  4.3.5.3  Centre versus batter outflow Outflow originating beneath the batters (batter outflow) and outflow originating  beneath the crown (centre outflow) has been estimated. This has been completed for the Type III test pile only, since a wetting front had not reached the base of the Type I test pile below the crown by the end of 2011. The basal liner of each test pile, partitioned into a main basal drain and a series of sub-lysimeters (i.e. BCLs), facilitated this estimate. For each year, the total volume of water from all BCLs underlying the crown was normalized by total BCL collection area and then multiplied by the total basal area underlying the crown to obtain an estimate of centre outflow. These estimates assume that the BCLs are representative of the entire basal area underlying the crown. The two aspects that add the greatest uncertainty to these estimates are 1) heterogeneity in flowpaths within the pile and 2) the amount of time that different regions of the pile remain thawed in a given year. Significant variation in outflow volumes between the BCLs in a given year illustrates the degree of heterogeneity in  116  flow paths within the Type III test pile. For example, in 2008 only 43 % of the BCL area recorded outflow (three 4 m by 4 m BCLs), which ranged from 187 L to 1211 L. Using temperature data from the deepest (11 m) thermistors in the uncovered test piles, Face 1 (near BCLs) remaining thawed for 20 days longer, on average, than Face 4 (Pham, in communication). The longer thaw time would have, potentially, facilitated outflow over a longer period of time each year at the BCLs than other basal regions underlying the crown. Table 4-5 summarizes the results for 2007 through 2011 at the Type III test pile. In 2007, 0 % of total outflow was centre outflow. This is supported by the lack of BCL outflow in 2007 and by TDR evidence showing residual matrix moisture content at 9 m depth. As a result, batter outflow accounted for all outflow in 2007. The arrival of wetting fronts at the base of the pile, below the crown, resulted in centre outflow accounting for approximately 16 % of total outflow in 2008. Centre outflow in 2009 and 2010 was much lower at 6 % and 3 % of total outflow respectively. This corresponded with no new wetting front arrivals at the 9 m depth TDR sensors in those years. Centre outflow in 2011 was the most of any year, both volumetrically and as a percent of total outflow, at 28 % of total outflow. These calculations show the extent to which the uncovered test piles are batter dominated systems; the basal area underlying the batters accounted for a minimum of 72 % of total outflow in 2011 to a maximum of 100 % of total outflow in 2007. This is a significant finding, since the proportion of the basal liner underlying the crown versus the batters is similar (55% batter and 45% crown).  117  4.4  Conclusions The freeze-thaw cycle, initial matrix wet-up, fluid flow, and outflow cycle at the  uncovered test piles were explored. The following conclusions are highlighted:  1. An active-zone develops through the summer months as a thaw front migrates inwards into the uncovered test piles. The freeze-thaw trend below 11 m was not observed due to thermistor placement; however, BCL outflow at the Type III test pile indicated that at least a portion of the base thawed for up to 2 months each year. As air temperatures drop below freezing each winter, deeper central locations freeze prior to shallower central locations due to the combined effects of air convection at the highly permeable toes and permafrost below the liner.  2. Differences in initial matrix wet-up times beneath the crowns of the uncovered test piles is attributed to applied rainfall events at the crown of the Type III test pile, which 1) provided additional water to the Type III test pile, and 2) likely resulted in increased porewater displacement within the matrix. It took approximately 25 months for the Type III test pile to wet-up to its base, and 62 months for the Type I test pile to wet-up to a depth of 9 m. These estimates include times when the test piles were frozen.  3. Wetting front velocities remained similar to initial estimates by Neuner et al. (2012) and were spatially uniform for rainfall intensities up to and including the 16.3 mm/d event in 2011. The minimum rainfall intensity necessary for triggering spatially 118  variable wetting front velocities during above field-capacity matrix conditions is between 17 mm/d and 30 mm/d at the uncovered test piles.  4. Outflow from the uncovered test piles occurred in an annual cycle that corresponded with the development of an active layer. Variations in the timing and magnitude of outflow over multiple years is a function of the timing of active zone formation, snow accumulation and melt, timing and magnitude of rainfall events, and antecedent moisture contents and wetting front locations held in storage over the previous winter. A positive correlation between outflow rate and sulfate/dissolved metals in outflow waters at both the active-zone lysimeters and the Type III basal collection lysimeters suggests that porewater displacement in the matrix is the dominant mechanism of flow in the active-zone lysimeters and below the crown of the Type III test pile.  5. Batter outflow makes up the majority of total outflow at the Type III test pile, constituting 100 %, 84 %, 94%, 97%, and 72 % of total outflow in 2007, 2008, 2009, 2010, and 2011 respectively. A wetting front did not reach the base of the Type I test pile by the end of 2011, resulting in batter outflow near 100 % of total outflow in all years. This is a significant finding, since the proportion of the basal liner underlying the crown versus the batters is similar (55% batter and 45% crown).  6. The design of the drainage scheme at the Type III test pile was superior to the scheme at the Type I test pile in a climate that induces annual freeze-thaw in waste-rock.  119  Placing perforated drainlines near the toes of the pile allowed water that reached the basal liner to flow through thawed waste-rock as the active-layer formed  The uncovered test piles are heterogeneous and represent a multi-permeability system. Unfortunately, only processes occurring in the matrix domain are observable with the current installed instrumentation. Temperature and outflow data indicate that the active-zone may develop into the entire extent of the uncovered test piles each year. This is very different from active-zone‎formation‎at‎DMMI’s‎80‎m‎stockpiles,‎where‎preliminary‎temperature‎data‎ from thermistors installed in boreholes indicate a maximum active-zone extent of 10-11 m depth below the crown. This difference in active-zone formation implies that while the hydrologic processes at the test piles can likely be applied to a larger stockpile, the seasonality and magnitude of discharge observed are unique to the scale of the test piles. The question of an appropriate REV for an experimental pile, in terms of fluid flow, still remains.  120  4.5  Figures  Figure ‎4-1  North-South cross-section of the Type I test pile, 2 m west of the centre line. TDR sensor  locations denoted by red dots. Adapted from Smith (2009)  121  Figure ‎4-2  (a): West-East cross-section of the Type III test pile, 2 m south of the centre line.  (b): West-East cross-section of the Type III test pile, 2 m north of the centre line. TDR sensor locations denoted by red dots. Adapted from Smith (2009)  122  Figure ‎4-3  Active zone lysimeter and air temperatures. The 0.6 m ECH 2O sensor was not recoverable after significant wire damage was sustained  at the end of September, 2009.  123  Figure ‎4-4  Comparison of 31S2 TDR apparent volumentric moisture content and 31S5 thermistor temperature.  124  Figure ‎4-5  Comparison of 12W2 TDR 9 m apparent volumentric moisture content and 12W5 thermistor 9 m temperature.  125  Figure ‎4-6  Apparent volumentric moisture content at the 31S2 TDR string (Type III test pile) for  2009  126  Figure ‎4-7  Volumetric moisture content at the 31S2 TDR string in the Type III test pile  127  Figure ‎4-8  Volumetric moisture content at the 33N2 TDR string in the Type III test pile  128  Figure ‎4-9  1 m depth TDR sensors in the Type III test pile  129  Figure ‎4-10  7 m depth TDR sensors in the Type III test pile  130  Figure ‎4-11  9 m depth TDR sensors in the Type III test pile  131  Figure ‎4-12  1 m depth TDR sensors in the Type III test pile from 2006 through 2007  Figure ‎4-13  7 m depth TDR sensors in the Type III test pile during 2007  132  Figure ‎4-14  Volumetric moisture content at the 11W2and 12W2 TDR strings in the Type I test pile  133  ~16 mm of outflow missed at Type III West.  Figure ‎4-15  2007 through 2011 cumulative annual outflow (normalized by surface collection area) at the active-zone lysimeters  134  Figure ‎4-16  2011 daily outflow at the Type I East (a) and Type I West (b) active-zone lysimeters  135  Figure ‎4-17  Type I test pile main basal drain outflow  136  Figure ‎4-18  Type III test pile main basal drain outflow  137  1  Figure ‎4-19  Outflow from the 3BNBlys2W/2E Type III basal collection lysimeters and apparent volumetric moisture contents along the 31S2 TDR  string  138  Figure ‎4-20  Outflow from the 3BNClys4E Type III basal collection lysimeter and apparent volumetric moisture contents along the 31S2 TDR  string  139  Figure ‎4-21  2011 Type III basal drain outflow  Figure ‎4-22  2011 3BNClys2W/2E outflow and 31S2 TDR apparent volumetric moisture contents  140  Figure ‎4-23  Type I and Type III basal drain outflow  141  4.6  Tables  Table ‎4-1  Dates for the onset of thawed and frozen conditions at the active zone lysimeters  Year 2007 2008 2009 2010 2011  Table ‎4-2  0.3 m (Thaw) May-21 May-31 May-15  0.3 m (Freeze) Oct-18 Oct-20 Oct-16 Oct-21  0.9 m (Thaw) May-28 June-10 May-27  0.9 m (Freeze) Oct-29 Oct-29 Nov-04  Active zone lysimeter outflow volumes  AZL  Flow Start  Flow Stop  TI West TI East TIII West TIII East  28/Aug/2007 25/Aug/2007  10/Oct/2007 10/Nov/2007  TI West TI East TIII West TIII East  17/Aug/2008 25/Aug/2008 27/Jun/2008 26/Jun/2008  21/Oct/2008 23/Oct/2008 23/Oct/2008 21/Oct/2008  TI West TI East TIII West TIII East  26/Jun/2009 26/Jun/2009 1/Jul/2009 28/Jun/2009  15/Oct/2009 15/Oct/2009 25/Oct/2009 25/Oct/2009  TI West TI East TIII West TIII East  21/Jun/2010 10/Jun/2010 10/Jun/2010 17/Jun/2010  19/Oct/2010 19/Oct/2010 21/Oct/2010 21/Oct/2010  Recorded Geochem. Outflow Sample (L) (L) 2007 No Flow No Flow 2008 2009 2010 131 16 116 18 68 18 71 17 2011  Total Outflow (L)  Total Outflow (mm)  36 31  18 15  297 194 203 202  83 54 97 97  66 22 23 38  18 6 11 18  147 134 84 88  41 37 40 42  142  309 86 283 79 146 24/Oct/2011 132 14 70 (86*) TIII West 5/Jun/2011 (180*) 5/Jun/2011 20/Oct/2011 170 15 185 88 TIII East *Values in parentheses include the estimated outflow missed in 2011 at the Type III West active-zone lysimeter TI West TI East  Table ‎4-3  Test Pile  4/Jun/2011 5/Jun/2011  20/Oct/2011 24/Oct/2011  295 268  14 15  Basal drain outflow volumes at the uncovered test piles  Flow Start  Flow Stop  Recorded Discharge (m3)  Interpolated Discharge (m3)  Total Discharge (m3)  0.3 40  unknown 70  unknown 110  50 130  1 20  51 150  9.4 100  0.5 17  10 117  44 213  0.03 0.06  44 213  83 176  0.02 0.04  83 176  2007 Type I Type III  1/May/2007 3/May/2007  Type I Type III  10/May/2008 2/May/2008  Type I Type III  22/May/2009 29/May/2009  Type I Type III  12/Apr/2010 20/Apr/2010  Type I Type III  9/May/2011 9/May/2011  unknown 1/Nov/2007 2008 25/Oct/2008 30/Oct/2008 2009 6/Oct/2000 30/Oct/2009 2010 10/Nov/2010 6/Nov/2010 2011 28/Oct/2011 4/Nov/2011  143  Table ‎4-4  Interpolated basal drain discharge at the uncovered test piles  Year  Interpolated Discharge Type I Test Pile unknown 1 0.5 0.03 0.02 Type III Test Pile 70 20 17 0.06 0.04  2007 2008 2009 2010 2011 2007 2008 2009 2010 2011  Table ‎4-5  Explanation Disconnected drainline Datalogger malfunction Datalogger malfunction Geochemical sampling Geochemical sampling Original TB overwhelmed Datalogger malfunction Datalogger malfunction Geochemical sampling Geochemical sampling  Centre versus batter outflow at the Type III test pile  Year 2007 2008 2009 2010 2011  Centre outflow Batter outflow (% of total outflow) (% of total outflow) 0% 100 % 16 % 84 % 6% 94 % 3% 97 % 28 % 72 %  144  Chapter 5: Conclusions and recommendations  5.1  Conclusions The principal aim of this research is to better understand fluid flow through mine  waste-rock at multiple scales in cold climates. The hydrologic response of two large-scale, experimental waste-rock piles (test piles) and a series of smaller-scale lysimeters (activezone lysimeters) were monitored. Both the test piles and the active-zone lysimeters were constructed on top of water collection systems in order to monitor outflows, and were instrumented with tensiometers, moisture content sensors, and thermistors, deployed within the fine grained fraction of the waste-rock. The dataset collected during this research represents an extensive fluid flow dataset in the field of mine waste-rock, and has been used to examine active-zone development, net infiltration, matrix wet-up, wetting front movement, and outflow at the test piles and active-zone lysimeters. The following conclusions are highlighted:  1. Above-freezing air temperatures result in an active-zone developing inwards into the test piles during the summer months, allowing net infiltration at the thawed surface and fluid flow through thawed regions to occur. Outflow from basal collection lysimeters at the base of the Type III test pile indicate that a portion of the 15 m high test piles thaw entirely to the base below the crown. This suggests that the active-zone may develop through the entire extent of the test piles each year. This is different from active-zone development at DDMIs 80 m high waste-rock piles. Preliminary temperature data indicate that the active-zone only extends 10-11 m below the crown of the stockpile  145  (Pham, 2012), suggesting that while the hydrologic processes at the test piles can be applied to the full-scale waste-rock pile, the seasonality and magnitude of discharge observed at the test piles are unique to that scale.  2. Infiltration into the test piles and active-zone lysimeters is restricted to periods when the surface of the waste-rock is thawed (typically May through October). Over a five year period the average percent net infiltration of natural rainfall into the active-zone lysimeters and crowns of the uncovered test piles was estimated to be 43%, 57%, 15%, 41%, and 58% of natural rainfall in 2007, 2008, 2009, 2010, and 2011, respectively. The significant variability in percent net infiltration over a five year period suggests that caution should be used when applying an average value for annual percent net infiltration. The FAO-PM method reasonably predicted net infiltration into the activezone lysimeters for multiple years when the value chosen for Ze (depth of evaporation layer) was calibrated to obtain a single best-fit value for net infiltration at the AZLs from 2008 through 2011. This shows that an indirect evapotranspiration computation developed for soils can predict net infiltration into waste-rock for multiple years using meteorological data and surface properties of waste-rock piles. However, the sensitivity of the FAO-PM method to the value chosen for Ze means that this method should be used with caution when assigning literature-values of Ze to waste-rock. Based on the FAO-PM estimates, net infiltration from small rainfall events (<5 mm/d) is restricted to periods early (May) and late in the rain-season (September to mid-October), while net infiltration in June, July, and August is restricted to larger rainfall events (>5 mm/d).  146  3. The initial wet-up of the matrix material below the crowns of the uncovered test piles was monitored using TDR sensors installed within matrix material along tip faces. Moisture content data from TDR sensors suggest that the difference in wet-up time between the Type I and III test piles was due to applied rainfall events at the crown of the Type III test pile, which 1) provided additional water to the Type III test pile, and 2) resulted in increased porewater displacement within the matrix compared to the equivalent volume of net infiltration at the Type I test pile. Including the periods when the test piles were frozen, it took approximately 25 months for the Type III test pile to wet-up to its base, and 62 months for the Type I test pile to wet-up to a depth of 9 m.  4. Within the matrix, wetting front velocities were approximately spatially uniform following all but the most intense rainfall events. Wetting front velocities calculated following rainfall events at the uncovered test piles suggest that the lowest intensity rainfall event that will trigger spatially variable wetting front velocities is somewhere between 17 mm/d and 29 mm/d. Spatially variable wetting front velocities were shown by Neuner et al. (2012) to be indicative of macropore flow, however, a positive correlation between outflow rate and solute/dissolved metal concentrations at the basal collection lysimeters and active-zone lysimeters suggest that the dominant outflow response at drain-depth was matrix flow. Therefore the dominant mechanism of fluid flow at the test piles and AZLs is porewater displacement in the matrix from propagating pressure waves, as opposed to preferential mechanisms such as macropore flow. The above conclusions apply only to the material underlying the crowns of the uncovered test piles.  147  5. Outflow from the uncovered test piles occurs in an annual cycle that corresponds with the development of the active-zone. The basal area underlying the crown and batters of the uncovered test piles is similar (55% batter and 45% crown); however, outflow data suggests that the piles remain batter dominated systems even after the matrix material below the crowns completely wet-up. Batter outflow was 100%, 84%, 94%, 97%, and 72% of total outflow in 2007, 2008, 2009, 2010, and 2011 respectively. The large difference in Type I and III test pile outflow cannot be fully attributed to rainfall events conducted at the Type III test pile and is likely a result of several factors, including differences in snowmelt infiltration on the batters, differences in drainage scheme, and differences in protective crush material overlying the basal drains.  6. TDR sensors were successfully used to monitor the active-zone development within the test piles, due to the different dielectric permittivities of ice and liquid water. Additionally, the correspondence of matrix wet-up (recorded by TDR sensors) and the initiation of outflow at the Type III basal collection lysimeters suggests that TDR sensors can be used to accurately monitor wet-up of the matrix fraction of waste-rock piles.  7. The active-zone lysimeters were instrumental in interpreting test pile behaviour, particularly in terms of net infiltration. Lysimeter-scale experiments should be adopted in future experiments, and effort put into constructing them so as to have similar climatic conditions at the upper surfaces of the lysimeters and any larger-scale experiments.  148  5.2  Recommendations The following recommendations are supplied as suggestions for continued study of  the test piles and as suggestions for future experimental waste-rock piles:  1. Tracer data from an experiment carried out at the crown of the Type III test pile in 2007 should be analysed in order to examine tracer velocities and the degree of preferential flow during high-intensity rainfall events.  2. Further work should be done to explain the large difference in annual outflow between the Type I and III test piles, including the presence/absence of a permanently frozen core and leaks in the main drain pipe.  3. Effort should be directed to measuring the snow depth at the batters of the test piles just prior to melt. Quantifying snowmelt infiltration into the batters of the test piles remains the largest missing component of the test pile water balance.  4. Effort should be directed to quantifying evaporation and infiltration at the batters of the test piles.  5. Consider replacing the current pressure-style flow-through cells at the basal collection lysimeters with gravity-fed cells to minimize data loss from leaks.  6. Implement a rigorous procedure for documenting geochemical sampling volumes at the active-zone lysimeters, in order to preserve the water balance calculation (completed).  149  7. For future field-scale experiments thermistors should be installed near the base of the pile, including along the liner, in order to accurately map which regions of the pile are frozen and thawed over time.  8. TDR sensors and thermistors should be co-located within the waste-rock piles to aid in the interpretation of fluid flow around the phase change of water.  9. To aid in the estimation of evaporation and net infiltration, moisture content sensors should be closely spaced within the upper 0.5 m of the waste-rock pile. Moisture content sensors should be installed throughout the entire depth of lysimeters to enable accurate storage estimates to be made.  10. To enable better monitoring of spatial and temporal variability in outflow, a contiguous grid of basal collection lysimeters should be constructed at the base of future experiments.  150  References  Allen, R.G., Pereira, L.S., Raes, D., Smith, M., 1998. Crop evapotranspiration: Guidelines for computing crop water requirements. FAO Irrigation and drainage paper 56, Food and Agriculture Organization of the United Nations, Rome, Italy. Amos, R.T., Blowes, D.W., Smith, L., Sego, D.C., 2009. Measurement of wind-induced pressure gradients in a waste rock pile. Vadose Zone Journal, v. 8, pp. 953-962. Bay, D., 2009. Hydrological and hydrogeochemical characteristics of neutral drainage from a waste rock test pile. M.Sc. thesis, University of British Columbia, Vancouver, Canada. Bailey, B.L., in progress. Ph.D. dissertation. University of Waterloo, Waterloo, Canada. Blowes, D.W., Ptacek, C.J., Jambor, J.L., Weisener, C.G., 2003. The geochemistry of acid mine drainage. In: Sherwood Lollar, B., Holland, H.D., and Turekian, K.K. (Eds.), Treatise on Geochemistry, v. 9, pp. 149-204. Brookes, R.H., Corey, A.T., 1964. Hydraulic properties of porous media. Hydrology Papers, 3, Colorado State University, Fort Collins. Burba, G., Anderson, D., 2010. A Brief Practical Guide to Eddy Covariance Flux Measurements: Principles and Workflow Examples for Scientific and Industrial Applications. LI-COR, Lincoln, USA. Carey, S.K., Barbour, S.L., Hendry, M.J., 2005. Evaporation from a waste-rock surface, Key Lake, Saskatchewan. Can. Geotech. J., v. 42, pp. 1189-1199. Chi, X., 2010. Characterizing low-sulfide instrumented waste-rock piles: Image grain-size analysis and wind-induced gas transport. M.Sc. thesis. University of Waterloo, Waterloo, Canada. Choudhury, B.J., 1997. Global Pattern of Potential Evaporation Calculated from the PenmanMonteith Equation Using Satellite and Assimilated Data. Remote Sens. Environ., v. 61, pp. 64-81. Fala, O., Molson, J., Aubertin, M., Bussière, B. 2005. Numerical modeling of flow and capillary barrier effects in unsaturated waste rock piles. Mine water and the environment, v. 24, pp. 172185. Fayer, M.J., Rockhold, M.L., Campbell, M.D., 1992. Hydrologic Modeling of Protective Barriers: Comparison of Field Data and Simulation Results. Soil Sci. Am. J., v. 56, pp. 690-700.  151  Flury, M., Fluhler, H., Jury, W.A., Leuenberger, J., 1994. Susceptibility of soils to preferential flow of water: a field study. Water Resources Research, v. 30, n. 7, pp. 1945 – 1954. Fujiyasu, Y., Fahey, M., and Newson, T.A., 2000. Field investigation of evaporation from freshwater tailings. Journal of Geotechnical and Geoenvironmental Engineering, v. 126, pp. 556–567. Gee, G.W., Hillel, D., 2006. Groundwater recharge in arid regions: Review and critique of estimation methods. Hydrological Processes Journal, v. 2(3) pp. 255-266. Gerke, H.H., van Genuchten, M. Th., 1993. A dual-porosity model for simulating the preferential movement of water and solutes in structured porous media. Water Resour. Res., v. 29, pp. 305–319. Golder Associates, 2008. Report on 2007 review of baseline climate and surface water hydrology for the Diavik Diamond Mine, prepared by Golder Associates, Burnaby, BC, submitted to Diavik Diamond Mines Inc., Yellowknife, NWT, Canada. Granger, R.J., Gray, D.M., 1989. Evaporation from natural nonsaturated surfaces. Journal of Hydrology, v. 111, pp. 21–29. Grismer, M.E., Orang, M., Snyder, R., Matyac, R., 2002. Pan evaporation to reference evapotranspiration conversion methods. J. Irrig. Drain Eng., v. 128, pp. 180-184. Hargreaves, G.H. and Z.A. Samani., 1985. Reference crop evapotranspiration from temperature. Applied Engrg. in Agr., Am. Soc. Agr. Engr, v. 1(2), pp. 96-99. Howell, T.A., Schneider, A.D., Jensen, M.E., 1991. History of lysimeter design and use for evapotranspiration measurements. In: Proceedings of the conference on lysimeters for evapotranspiration and environmental measurements, Honolulu. Hydrological Services PTY LTD, 2003. Instruction manual: Evaporation pan class A. Prepared by Hydrological Services PTY LTD, Australia, Issue 1, CAEP100-01. Lefebvre, R., Hockley, D., Smolensky, J., Gélinas, P., 2001. Multiphase transfer processes in waste rock piles producing acid mine drainage: 1: Conceptual model and system characterization. Journal of Contaminant Hydrogeology, v. 52, pp. 137-164. Mualem, Y., 1976. A new model for predicting the hydraulic conductivity of unsaturated porous media. Water Resources Research, v. 12, n. 3, p. 513. Momeyer, S., in progress. M.Sc. thesis. University of British Columbia, Vancouver, Canada. Monteith, J.L., 1965. Evaporation and Environment. 19th Symposia of the Society for Experimental Biology, University Press, Cambridge, pp. 205–234. 152  Neuner, M., 2009. Water flow through unsaturated mine waste rock in a region of permafrost. M.Sc. thesis, University of British Columbia, Vancouver, Canada. Neuner, M., Smith, L., Blowes, D.W., Sego, D.C., Smith, L.J.D., Fretz, N., Gupton, M., accepted. The Diavik Waste Rock Project: Water flow through mine waste in a permafrost terrain, Applied Geochemistry (2012). Newson, T.A., Fahey, M., 2003. Measurement of evaporation from saline tailings storages. Engineering Geology, v. 70(3-4), pp. 217-233. Nichol, C.F., 2002. Transient flow and transport in unsaturated heterogeneous media: Field experiments in mine waste rock. Ph.D. dissertation. University of British Columbia, Vancouver, Canada. Nichol, C.F., Smith, L., Beckie, R., 2005. Field scale experiments of unsaturated flow and solute transport in a heterogeneous porous medium. Water Resources Research, v. 41. O’Kane,‎M.,‎Wilson,‎G.W.,‎Barbour,‎S.L.,‎1998.‎Instrumentation‎and‎monitoring‎of‎an‎ engineered soil cover system for mine waste rock. Canadian Geotechnical Journal, v. 35(5), pp. 828-846. Penman, H.L., 1948. Natural evaporation from open water, bare soil, and grass. Proceedings of the Royal Society of London, A193, pp. 120–145. Pham, H., 2012. Heat transfer in waste-rock piles constructed in a continuous permafrost region. Ph.D. dissertation. University of Alberta, Edmonton, Canada. Rasmussen T.C., Baldwin Jr., R.H., Dowd, J.F., Williams, A.G., 2000, Tracer vs. pressure wave velocities through unsaturated saprolite, Soil Sci. Soc. of Am. J., v. 64(1) pp. 75-85. Smith, L., Beckie, R., 2003. Hydrologic and geochemical transport processes in mine waste rock. In: Jambor, J.L, Blowes, D.W., Ritchie, A.I.M. (Eds.), Environmental Aspects of Mine Wastes. Mineral. Assoc. Can. Short Course, 31, pp. 51-72. Smith, L.J.D., 2009. Building and characterizing low sulfide instrumented waste rock piles: Pile design and construction, particle size and sulfur characterization, and initial geochemical response. M.Sc. thesis. University of Waterloo, Waterloo, Canada. Smith, L.J.D., Bailey, B.L., Blowes, D.W., Jambor, J.L, Smith., L., Sego, D.C., in press. The Diavik waste rock project: Initial geochemical response from a low sulfide waste rock pile. Applied Geochemistry (2012). Stockwell, J., Smith, L., Jambor, J.L., Beckie, R., 2006. The relationship between fluid flow and mineral weathering in heterogeneous unsaturated porous media: A physical and geochemical characterization of a waste-rock pile. Appl. Geochem., v. 21, pp. 1347-1361. 153  Thornthwaite, C.W., Holzman, B. 1939. The determination of evaporation from land and water surfaces. Monthly Weather Review, v. 67, pp. 4–11. Topp, G.C., Davis, J.L., Annan, A.P., 1980. Electromagnetic determination of soil water content: Measurement in coaxial transmission lines. Water Resources Research, v. 16(3), pp. 574-582. van Genuchten, M. Th., 1980. A closed-form equation for predicting the hydraulic conductivity of unsaturated soils, Soil Sci. Soc. Am. J., v. 44, pp. 892-898. van Genuchten, M. Th., Wierenga, P.J., 1976. Mass transfer studies in sorbing porous media. I. Analytical solutions. Soil Sci. Soc. Am. J., v. 40, pp. 473–481. Wagner, K., Smith, L., Beckie, R., 2006. Hydrogeochemical characterization of effluent from mine waste rock, Cluff Lake, Saskatchewan. In: Proceedings of the 7th International Conference on Acid Rock Drainage, St. Louis, MO. Wilson, G.W., Fredlund, D.G., Barbour, S.L., 1994. Coupled soil-atmosphere modelling for soil evaporation. Canadian Geotechnical Journal, v. 31(2), pp. 151-161. Yanful, E.K., Choo, L.P., 1997. Measurement of evaporative fluxes from candidate cover soils. Canadian Geotechnical Journal, v. 34(3), pp. 447-459. Yanful, E.K., Mousavi, S.M., De Souza, L-P., 2006. A numerical study of soil cover performance. Journal of Environmental Management, v. 81(1), pp. 72-92 Yazdani, J., Barbour, L., Wilson, G.W., 2000. Soil water characteristic curve for mine waste rock containing coarse material. In: Proceedings of the 6th Environmental Engineering Specialty Conference of the Canadian Society of Civil Engineers, London, ON, Canada.  154  Appendices  Appendix A Hydrology dataset The complete hydrology dataset from 2007 through 2011 is documented here. Experimental set-up and design, which includes the locations of experiments and instruments within the Test Piles Research Area, was presented in Chapter 2.  A.1  Rainfall At the Test Piles Research Area, natural rainfall was calculated as the average rainfall  from four rain gauge tipping buckets at the crowns of the test piles, while applied rainfall and tracer events were measured by placing a grid of cups at the ground surface within each event’s‎radius‎of‎influence‎(Neuner,‎2009;‎Momeyer,‎in‎progress).‎Figure‎A-1 presents natural rainfall at the Test Piles Research Area from 2007 through 2011. Figures A-2 to A-9 present natural and applied rainfall at individual areas within the Test Piles Research Area, from 2007 through 2011. Rainfall in Figures A-1 to A-9 is presented as daily rainfall, in millimetres. Table A-1 summarizes annual natural rainfall at the Test Piles Research area, and Table A-2 summarizes annual rainfall (including applied events) at individual areas within the Test Piles Research Area. Tables A-3 and A-4 document applied rainfall events at the Type III test pile and Type III active-zone lysimeters, respectively.  155  Figure ‎A-1  2007 through 2011 natural rainfall at the Test Piles Research Area (all test piles and active-zone lysimeters)  156  Figure ‎A-2  2007 natural rainfall at the Type I test pile, Type III test pile batters, Covered test pile,  and Type I active-zone lysimeters  Figure ‎A-3  2007 natural and applied rainfall at the crown of the Type III test pile  157  Figure ‎A-4  2007 natural and applied rainfall at the Type III active-zone lysimeters  Figure ‎A-5  2008 natural rainfall at the Type I test pile, Type III test pile, Covered test pile, and  Type I active-zone lysimeters  158  Figure ‎A-6  2008 natural and applied rainfall at the Type III active-zone lysimeters  Figure ‎A-7  2009 natural rainfall for the Test Piles Research Area  159  Figure ‎A-8  2010 natural rainfall for the Test Piles Research Area  Figure ‎A-9  2011 natural rainfall for the Test Piles Research Area  160  Table ‎A-1  Natural rainfall measured at the crowns of the test piles  Year 2007 2008 2009 2010 2011  Table ‎A-2  Total annual rainfall at the test piles and active-zone lysimeters (AZLs)  Type III test pile crown (mm) 152.7 154.4 74.0 97.5 145.5  Year 2007 2008 2009 2010 2011  Table ‎A-3  Type III AZLs (mm) 183.7 179.9 74.0 97.5 145.5  Type I test pile Type III test pile batters Type I AZLs (mm) 92.4 154.4 74.0 97.5 145.5  Applied rainfall at the crown of the Type III test pile (area of influence = 20mx30m)  Date 20/Sep/2006 24/Sep/2006 26/Sep/2006 17/Aug/2007 4/Sep/2007 13/Sep/2007  Table ‎A-4  Natural rainfall (mm) 92.4 154.4 74.0 97.5 145.5  Applied rainfall at Type III test pile crown (mm) 24.2 19.0 14.7 16.1 15.0 29.2  Rainfall intensity (mm/hr) 10.2 9.5 9.8 7.6 8.6 7.7  Tracer D2O Cl & Br-  Applied rainfall at the Type III active-zone lysimeters (AZLs)  Date 4/Aug/2007 5/Aug/2007 6/Aug/2007 1/Sep/2007 19/Sep/2007  Applied rainfall at Type III AZLs (mm) 23.1 27.7 14.1 14.8 11.6  Rainfall intensity (mm/hr) 8.2 12.3 8.2 8.4 10.8  Tracer 161  6/Aug/2008 12/Aug/2008 A.2 Air temperature  10.6 14.9  15.2 16.3  Cl-  Figures A-10 to A-15 present air temperature, representative of conditions at the Test Piles Research Area, from 2007 through 2011. Air temperature was collected at the DDMI meteorological station, approximately 1 km from the Test Piles Research Area, and is presented hourly in degrees Celsius.  162  Figure ‎A-10  2007 through 2011 air temperature at the Test Piles Research Area  163  Figure ‎A-11  2007 air temperature at the Test Piles Research Area  Figure ‎A-12  2008 air temperature at the Test Piles Research Area  164  Figure ‎A-13  2009 air temperature at the Test Piles Research Area  Figure ‎A-14  2010 air temperature at the Test Piles Research Area  165  Figure ‎A-15  2011 air temperature at the Test Piles Research Area  166  A.3  Type I test pile The Type I test pile dataset consists of volumetric moisture contents from TDR  sensors and outflow from the main basal drain and basal collection lysimeters. Figures A-16 to A-21 present apparent volumetric moisture contents at the TDR sensors from 2007 through 2011. Figure A-22 presents daily and annual outflow from the main basal drain, from 2007 through 2011. Figures A-23 to A-27 present basal drain outflow in each year, individually. Figures A-28 to A-37 present outflow from the basal collection lysimeters, from 2007 through 2011. Tables A-5 and A-6 summarize annual outflow volumes from the main basal drain and basal collection lysimeters, respectively. .  167  Figure ‎A-16  2007 through 2011 apparent volumetric moisture content at the Type I test pile  168  Figure ‎A-17  2007 apparent volumetric moisture content at the Type I test pile  Figure ‎A-18  2008 apparent volumetric moisture content at the Type I test pile  169  Figure ‎A-19  2009 apparent volumetric moisture content at the Type I test pile  Figure ‎A-20  2010 apparent volumetric moisture content at the Type I test pile  170  Figure ‎A-21  2011 apparent volumetric moisture content at the Type I test pile  171  Disconnected drain pipe. Unknown volume of water lost.  Figure ‎A-22  2007 through 2011 basal drain outflow at the Type I test pile  172  Disconnected drain pipe. Unknown volume of water lost.  Figure ‎A-23  2007 basal drain outflow at the Type I test pile  Figure ‎A-24  2008 basal drain outflow at the Type I test pile  173  Figure ‎A-25  2009 basal drain outflow at the Type I test pile  Figure ‎A-26  2010 basal drain outflow at the Type I test pile  174  Figure ‎A-27  2011 basal drain outflow at the Type I test pile  175  Figure ‎A-28  2007 through 2011 1BWBlys Cluster basal collection lysimeter outflow at the Type I test pile  176  Figure ‎A-29  2009 1BWBlys Cluster basal collection lysimeter outflow at the Type I test pile  Figure ‎A-30  2010 1BWBlys Cluster basal collection lysimeter outflow at the Type I test pile  177  Figure ‎A-31  2007 through 2011 1BWClys Cluster basal collection lysimeter outflow at the Type I test pile  178  Figure ‎A-32  2010 1BWClys Cluster basal collection lysimeter outflow at the Type I test pile  179  Figure ‎A-33  2007 through 2011 1BEClys Cluster basal collection lysimeter outflow at the Type I test pile. Note the scale.  180  Figure ‎A-34  2008 1BEClys Cluster basal collection lysimeter outflow at the Type I test pile  Figure ‎A-35  2009 1BEClys Cluster basal collection lysimeter outflow at the Type I test pile  181  Figure ‎A-36  2010 1BEClys Cluster basal collection lysimeter outflow at the Type I test pile  Figure ‎A-37  2011 1BEClys Cluster basal collection lysimeter outflow at the Type I test pile  182  Table ‎A-5  Basal drain outflow volumes at the Type I test pile  Year  Flow Start  Flow Stop  2007 2008 2009 2010 2011  1/May/2007 10/May/2008 22/May/2009 12/Apr/2010 9/May/2011  unknown 25/Oct/2008 6/Oct/2000 10/Nov/2010 28/Oct/2011  Table ‎A-6  Lysimeter Cluster  Recorded Discharge (m3) 0.3 50 9.4 44 83  Interpolated Discharge (m3) unknown 1 0.5 0.03 0.02  Total Discharge (m3) unknown 51 10 44 83  Basal collection lysimeter outflow volumes at the Type I test pile  Flow Start  Flow Stop  Recorded Discharge (L)  Interpolated Discharge (L)  Total Discharge (L)  0 -  1.5 -  1.5 3.5 18  -  -  62 1 123  91 55 172.5  8 1 4.5  99 56 177  No Flow No Flow 56  5  61  2007 No Flow  All 2008 1BWB 1BWC 1BEC  27/Aug/2008 29/Jun/2008  1BWB 1BWC 1BEC  7/Jul/2009 7/Jul/2009 7/Jul/2009  1BWB 1BWC 1BEC  3/Jul/2010 27/Jun/2010 1/Jun/2010  1BWB 1BWC 1BEC  14/May/2011  3/Sep/2008 6/Sep/2008 2009 29/Sep/2009 10/Jul/2009 29/Sep/2009 2010 18/Aug/2010 2/Aug/2010 16/Aug/2010 2011  21/May/2010  183  A.4  Type III test pile The Type III test pile dataset consists of volumetric moisture contents from TDR  sensors, outflow from the main basal drains and basal collection lysimeters, and hydraulic head from tensiometers at the crown of the pile. Figures A-38 to A-61 present apparent volumetric moisture contents at the TDR sensors from 2006 through 2011. Figure A-62 presents daily and annual outflow at the North and South (summed together) main basal drains, from 2007 through 2011. Figures A-63 to A-67 present basal drain outflow in each year, individually. Figure A-68 compares outflow at the North and South basal drains for 2010 and 2011. Figures A-69 to A-88 present outflow from the basal collection lysimeters from 2007 through 2011. Tables A-7 and A-8 summarize annual outflow volumes from the main basal drain and basal collection lysimeters, respectively. Figures A-89 and A-90 present hydraulic head calculated at tensiometers from 2010 through 2011.  184  Figure ‎A-38  2006 through 2011 apparent volumetric moisture content along TDR-string 31N2 at the Type III test pile  185  Figure ‎A-39  2007 apparent volumetric moisture content along TDR-string 31N2 at the Type III test  pile  Figure ‎A-40  2008 apparent volumetric moisture content along TDR-string 31N2 at the Type III test  pile  186  Figure ‎A-41  2009 apparent volumetric moisture content along TDR-string 31N2 at the Type III test  pile  Figure ‎A-42  2010 apparent volumetric moisture content along TDR-string 31N2 at the Type III test  pile  187  Figure ‎A-43  2009 apparent volumetric moisture content along TDR-string 31N2 at the Type III test  pile  188  Figure ‎A-44  2006 through 2011 apparent volumetric moisture content along TDR-string 31S2 at the Type III test pile  189  Figure ‎A-45  2007 apparent volumetric moisture content along TDR-string 31S2 at the Type III test  pile  Figure ‎A-46  2008 apparent volumetric moisture content along TDR-string 31S2 at the Type III test  pile  190  Figure ‎A-47  2009 apparent volumetric moisture content along TDR-string 31S2 at the Type III test  pile  Figure ‎A-48  2010 apparent volumetric moisture content along TDR-string 31S2 at the Type III test  pile  191  Figure ‎A-49  2011 apparent volumetric moisture content along TDR-string 31S2 at the Type III test  pile  192  Figure ‎A-50  2006 through 2011 apparent volumetric moisture content along TDR-string 33N2 at the Type III test pile  193  Figure ‎A-51  2007 apparent volumetric moisture content along TDR-string 33N2 at the Type III test  pile  Figure ‎A-52  2008 apparent volumetric moisture content along TDR-string 33N2 at the Type III test  pile  194  Figure ‎A-53  2009 apparent volumetric moisture content along TDR-string 33N2 at the Type III test  pile  Figure ‎A-54  2010 apparent volumetric moisture content along TDR-string 33N2 at the Type III test  pile  195  Figure ‎A-55  2011 apparent volumetric moisture content along TDR-string 33N2 at the Type III test  pile  196  Figure ‎A-56  2006 through 2011 apparent volumetric moisture content along TDR-string 33S2 at the Type III test pile  197  Figure ‎A-57  2007 apparent volumetric moisture content along TDR-string 33S2 at the Type III test  pile  Figure ‎A-58  2008 apparent volumetric moisture content along TDR-string 33S2 at the Type III test  pile  198  Figure ‎A-59  2009 apparent volumetric moisture content along TDR-string 33S2 at the Type III test  pile  Figure ‎A-60  2010 apparent volumetric moisture content along TDR-string 33S2 at the Type III test  pile  199  Figure ‎A-61  2011 apparent volumetric moisture content along TDR-string 33S2 at the Type III test  pile  200  Tipping bucket issues. ~70 m3 of water lost.  Figure ‎A-62  2007 through 2011 basal drain outflow at the Type III test pile (North and South drains combined)  201  Tipping bucket issues. ~70 m3 of water lost.  Figure ‎A-63  2007 basal drain outflow at the Type III test pile (North and South drains combined)  Figure ‎A-64  2008 basal drain outflow at the Type III test pile (North and South drains combined)  202  Figure ‎A-65  2009 basal drain outflow at the Type III test pile (North and South drains combined)  Figure ‎A-66  2010 basal drain outflow at the Type III test pile (North and South drains combined)  203  Figure ‎A-67  2011 basal drain outflow at the Type III test pile (North and South drains combined)  204  Figure ‎A-68 2010 and 2011comparison of the North and South basal drains at the Type III test pile  205  Figure ‎A-69  2007 through 2011 3BNBlys2W/2E basal collection lysimeter outflow at the Type III test pile  206  Figure ‎A-70  2009 3BNBlys2W/2E basal collection lysimeter outflow at the Type III test pile  Figure ‎A-71  2010 3BNBlys2W/2E basal collection lysimeter outflow at the Type III test pile  207  Figure ‎A-72  2007 through 2011 3BNBlys4W basal collection lysimeter outflow at the Type III test pile  208  Figure ‎A-73  2009 3BNBlys4W basal collection lysimeter outflow at the Type III test pile  Figure ‎A-74  2010 3BNBlys4W basal collection lysimeter outflow at the Type III test pile  209  Figure ‎A-75  2011 3BNBlys4W basal collection lysimeter outflow at the Type III test pile  210  Figure ‎A-76  2011 3BNClys2W/2E basal collection lysimeter outflow at the Type III test pile  211  Figure ‎A-77  2011 3BNClys2W/2E basal collection lysimeter outflow at the Type III test pile  212  Figure ‎A-78  2007 through 2011 3BNClys4W basal collection lysimeter outflow at the Type III test pile  213  Figure ‎A-79  2008 3BNClys4W basal collection lysimeter outflow at the Type III test pile  214  Figure ‎A-80  2007 through 2011 3BNClys4E basal collection lysimeter outflow at the Type III test pile  215  Figure ‎A-81  2008 3BNClys4E basal collection lysimeter outflow at the Type III test pile  Figure ‎A-82  2009 3BNClys4E basal collection lysimeter outflow at the Type III test pile  216  Figure ‎A-83  2010 3BNClys4E basal collection lysimeter outflow at the Type III test pile  Figure ‎A-84  2011 3BNClys4E basal collection lysimeter outflow at the Type III test pile  217  Figure ‎A-85  2007 through 2011 3BSClys4E basal collection lysimeter outflow at the Type III test pile  218  Figure ‎A-86  2008 3BSClys4E basal collection lysimeter outflow at the Type III test pile  Figure ‎A-87  2009 3BSClys4E basal collection lysimeter outflow at the Type III test pile  219  Figure ‎A-88  2007 through 2011 3BSClys4E basal collection lysimeter outflow at the Type III test pile  220  Figure ‎A-89  2010 hydraulic head calculated from tensiometers at the Type III test pile crown  Figure ‎A-90  2011 hydraulic head calculated from tensiometers at the Type III test pile crown  221  Table ‎A-7  Basal drain outflow volumes at the Type III test pile  Year  Flow Start  Flow Stop  2007 2008 2009 2010 2011  3/May/2007 2/May/2008 29/May/2009 20/Apr/2010 9/May/2011  1/Nov/2007 30/Oct/2008 30/Oct/2009 6/Nov/2010 4/Nov/2011  Table ‎A-8  Recorded Discharge (m3) 40 130 100 213 176  Interpolated Discharge (m3) 70 20 17 0.06 0.04  Total Discharge (m3) 110 150 117 213 176  Basal collection lysimeter outflow volumes at the Type III test pile  BCL  Flow Start  Flow Stop  Recorded Discharge (L)  Interpolated Discharge (L)  Total Discharge (L)  -  629 1211  -  187  -  86 307  -  1 228  -  10  2007 No Flow  All Years 2008 3BNBlys2W/2E 3BNBlys4W 3BNBlys4E 3BNClys2W/2E 6/Sep/2008 3BNClys4W 16/Aug/2008 3BNClys4E 3BSClys4W 31/Aug/2008 3BSClys4E 3BSClys2W/2E 3BNBlys2W/2E 3BNBlys4W 3BNBlys4E 3BNClys2W/2E 3BNClys4W 3BNClys4E 3BSClys4W 3BSClys4E 3BSClys2W/2E  4/Nov/2008 4/Nov/2008 4/Nov/2008  6/Jul/2009 7/Jul/2009  2009 7/Oct/2009 22/Oct/2009  7/Jul/2009 7/Jul/2009  10/Oct/2009 17/Oct/2009  7/Jul/2009  11/Oct/2009  No Flow No Flow No Flow No Flow No Flow No Flow No Flow No Flow No Flow No Flow  222  3BNBlys2W/2E 3BNBlys4W 3BNBlys4E 3BNClys2W/2E 3BNClys4W 3BNClys4E 3BSClys4W 3BSClys4E 3BSClys2W/2E  27/Jun/2010 6/Aug/2010  2010 20/Oct/2010 9/Oct/2010  29/Sep/2010  24/Oct/2010  3BNBlys2W/2E 16/Oct/2011 15/Aug/2011 3BNBlys4W 3BNBlys4E 3BNClys2W/2E 15/Sep/2011 3BNClys4W 1/Sep/2011 3BNClys4E 3BSClys4W 26/Sep/2011 3BSClys4E 3BSClys2W/2E  2011 21/Oct/2011 11/Nov/2011 13/Nov/2011 11/Nov/2011 13/Nov/2011  135 482 No Flow No Flow No Flow No Flow No Flow No Flow  3 11  138 493  -  3  4 952 No Flow 1096 No Flow 1601 No Flow 430 No Flow  1 11  5 963  9  1105  10  1611  6  436  223  A.5  Covered test pile The Covered test pile dataset consists of volumetric moisture contents from TDR  sensors, outflow from the main basal drain, hydraulic head from tensiometers at the crown of the pile, and temperature and volumetric moisture contents from ECH2O sensors. Figures A91 to A-105 present apparent volumetric moisture contents at the TDR sensors from 2008 through 2011. Figure A-106 presents daily and annual outflow at the main basal drain, from 2007 through 2011. Figures A-107 to A-110 present basal drain outflow in each year, individually. Table A-9 summarizes annual outflow volumes from the main basal drain. Figures A-111 and A-112 present hydraulic head calculated at tensiometers from 2010 through 2011. Figures A-113 to A-118 present temperature and volumetric moisture contents from ECH2O sensors within the till and Type III material from 2009 through 2010.  224  Figure ‎A-91  2008 through 2011 apparent volumetric moisture content along TDR-string C2E2 at the Covered test pile  225  Figure ‎A-92  2008 apparent volumetric moisture content along TDR-string C2E2 at the Covered test  pile  Figure ‎A-93  2009 apparent volumetric moisture content along TDR-string C2E2 at the Covered test  pile  226  Figure ‎A-94  2010 apparent volumetric moisture content along TDR-string C2E2 at the Covered test  pile  Figure ‎A-95  2011 apparent volumetric moisture content along TDR-string C2E2 at the Covered test  pile  227  Figure ‎A-96  2008 through 2011 apparent volumetric moisture content along TDR-string C2W2 at the Covered test pile  228  Figure ‎A-97  2008 apparent volumetric moisture content along TDR-string C2W2 at the Covered test  pile  Figure ‎A-98  2009 apparent volumetric moisture content along TDR-string C2W2 at the Covered test  pile  229  Figure ‎A-99  2010 apparent volumetric moisture content along TDR-string C2W2 at the Covered test  pile  Figure ‎A-100  2011 apparent volumetric moisture content along TDR-string C2W2 at the Covered test  pile  230  Figure ‎A-101  2008 through 2011 apparent volumetric moisture content along TDR-string C3E2 at the Covered test pile  231  Figure ‎A-102  2008 apparent volumetric moisture content along TDR-string C3E2 at the Covered test  pile  Figure ‎A-103  2009 apparent volumetric moisture content along TDR-string C3E2 at the Covered test  pile  232  Figure ‎A-104  2010 apparent volumetric moisture content along TDR-string C3E2 at the Covered test  pile  Figure ‎A-105  2011 apparent volumetric moisture content along TDR string C3E2 at the Covered test  pile  233  Figure ‎A-106  2007 through 2011 basal drain outflow at the Covered test pile  234  Figure ‎A-107  2007/2008 season basal drain outflow at the Covered test pile  Figure ‎A-108  2008/2009 season basal drain outflow at the Covered test pile  235  Figure ‎A-109  2009/2010 season basal drain outflow at the Covered test pile  236  Figure ‎A-110  2009/2011/2012 season basal drain outflow at the Covered test pile. Outflow still occurring after truncation in December, 2011.  237  Figure ‎A-111  2010 hydraulic head calculated from tensiometers at the Covered test pile crown  Figure ‎A-112  2011 hydraulic head calculated from tensiometers at the Covered test pile crown  238  Figure ‎A-113  2009 temperature and volumetric moisture content (VMC) within the Type III material  and till at the Covered test pile at STA 0+94  Figure ‎A-114  2010 temperature and volumetric moisture content (VMC) within the Type III material  and till at the Covered test pile at STA 0+94  239  Figure ‎A-115  2009 temperature and volumetric moisture content (VMC) at the top of the till layer at  the Covered test pile, on the batters downslope of the crown  Figure ‎A-116  2010 temperature and volumetric moisture content (VMC) at the top of the till layer at  the Covered test pile, on the batters downslope of the crown  240  Figure ‎A-117  2009 temperature and volumetric moisture content (VMC) at the bottom of the till layer  at the Covered test pile, on the batters downslope of the crown  Figure ‎A-118  2010 temperature and volumetric moisture content (VMC) at the bottom of the till layer  at the Covered test pile, on the batters downslope of the crown  241  Table ‎A-9  Basal drain outflow volumes at the Covered test pile  Recorded Interpolated Total Year Flow Start Flow Stop Discharge Discharge Discharge 3 3 (m ) (m ) (m3) 2007/08 8/Sep/2007 10/Jan/2008 0.5 0 0.5 2008/09 26/Aug/2008 10/Jul/2009 55 120 175 2009/10 6/Aug/2009 8/Jun/2010 30 0.1 30 2010/11/12 11/Jul/2010 * 296* 19 325* *No stoppage in outflow from 2010/11 season to 2011/12 season. Still flowing after truncation in dataset presented here.  242  A.6  Active-zone lysimeters (AZLs) The AZL dataset consists of temperature and volumetric moisture contents from  ECH2O sensors, hydraulic head from tensiometers, and outflow from each AZL. Figures A119 to A-124 present temperature and volumetric moisture contents from ECH2O sensors located adjacent to the active-zone lysimeters. Figures A-125 and A-126 present hydraulic head calculated at tensiometers adjacent to the active-zone lysimeters. The footprint from applied rainfall events to the Type III AZLs covered the ground surface above the ECH2O sensors and tensiometers. Therefore, in 2007 and 2008 the ECH2O sensor and tensiometer response represents conditions at the Type III AZLs only. Figures A-127 to A-148 present daily outflow from the active-zone lysimeters, while Figures A-149 to A-154 present cumulative annual outflow from the active-zone lysimeters. Cumulative annual outflow has been normalized by surface collection area at the AZLs, to facilitate a better comparison between the Type I and Type III AZLs. Table A-10 summarizes annual outflow volumes at the active-zone lysimeters.  243  Figure ‎A-119  2007 through 2011 temperature and volumetric moisture content (VMC) from ECH 2O sensors at the active zone-lysimeters  244  Figure ‎A-120  2007 temperature and volumetric moisture content (VMC) from ECH2O sensors at the  active zone-lysimeters  Figure ‎A-121  2008 temperature and volumetric moisture content (VMC) from ECH 2O sensors at the  active zone-lysimeters  245  Figure ‎A-122  2009 temperature and volumetric moisture content (VMC) from ECH 2O sensors at the  active zone-lysimeters  Figure ‎A-123  2010 temperature and volumetric moisture content (VMC) from ECH2O sensors at the  active zone-lysimeters  246  Figure ‎A-124  2011 temperature and volumetric moisture content (VMC) from ECH 2O sensors at the  active zone-lysimeters  247  Figure ‎A-125  2010 hydraulic head calculated from tensiometers at the active zone-lysimeters  Figure ‎A-126  2011 hydraulic head calculated from tensiometers at the active zone-lysimeters  248  Figure ‎A-127  2007 through 2011 outflow at the Type I East active-zone lysimeter  249  Figure ‎A-128  2008 outflow at the Type I East active-zone lysimeter  Figure ‎A-129  2009 outflow at the Type I East active-zone lysimeter  250  Figure ‎A-130  2010 outflow at the Type I East active-zone lysimeter  Figure ‎A-131  2011 outflow at the Type I East active-zone lysimeter  251  Figure ‎A-132  2007 through 2011 outflow at the Type I West active-zone lysimeter  252  Figure ‎A-133  2008 outflow at the Type I West active-zone lysimeter  Figure ‎A-134  2009 outflow at the Type I West active-zone lysimeter  253  Figure ‎A-135  2010 outflow at the Type I West active-zone lysimeter  Figure ‎A-136  2011 outflow at the Type I West active-zone lysimeter  254  Figure ‎A-137  2007 through 2011 outflow at the Type III East active-zone lysimeter  255  Figure ‎A-138  2007 outflow at the Type III East active-zone lysimeter  Figure ‎A-139  2008 outflow at the Type III East active-zone lysimeter  256  Figure ‎A-140  2009 outflow at the Type III East active-zone lysimeter  Figure ‎A-141  2010 outflow at the Type III East active-zone lysimeter  257  Figure ‎A-142  2011 outflow at the Type III East active-zone lysimeter  258  Figure ‎A-143  2007 through 2011 outflow at the Type III West active-zone lysimeter  259  Figure ‎A-144  2007 outflow at the Type III West active-zone lysimeter  Figure ‎A-145  2008 outflow at the Type III West active-zone lysimeter  260  Figure ‎A-146  2009 outflow at the Type III West active-zone lysimeter  Figure ‎A-147  2010 outflow at the Type III West active-zone lysimeter  261  Figure ‎A-148  2011 outflow at the Type III West active-zone lysimeter  262  ~16 mm of outflow missed at Type III West.  Figure ‎A-149  2007 through 2011 cumulative annual outflow (normalized by surface collection area) at the active-zone lysimeters  263  Figure ‎A-150  2007 cumulative annual outflow (normalized by surface collection area) at the active-  zone lysimeters  Figure ‎A-151  2008 cumulative annual outflow (normalized by surface collection area) at the active-  zone lysimeters  264  Figure ‎A-152  2009 cumulative annual outflow (normalized by surface collection area) at the active-  zone lysimeters  Figure ‎A-153  2010 cumulative annual outflow (normalized by surface collection area) at the active-  zone lysimeters  265  ~16 mm of outflow missed at Type III West.  Figure ‎A-154  2011 cumulative annual outflow (normalized by surface collection area) at the active-  zone lysimeters  266  Table ‎A-10  Basal drain outflow volumes at the Covered test pile  AZL  Flow Start  Flow Stop  TI West TI East TIII West TIII East  28/Aug/2007 25/Aug/2007  10/Oct/2007 10/Nov/2007  TI West TI East TIII West TIII East  17/Aug/2008 25/Aug/2008 27/Jun/2008 26/Jun/2008  21/Oct/2008 23/Oct/2008 23/Oct/2008 21/Oct/2008  TI West TI East TIII West TIII East  26/Jun/2009 26/Jun/2009 1/Jul/2009 28/Jun/2009  15/Oct/2009 15/Oct/2009 25/Oct/2009 25/Oct/2009  TI West TI East TIII West TIII East  21/Jun/2010 10/Jun/2010 10/Jun/2010 17/Jun/2010  19/Oct/2010 19/Oct/2010 21/Oct/2010 21/Oct/2010  4/Jun/2011 5/Jun/2011  20/Oct/2011 24/Oct/2011  Recorded Geochem. Outflow Sample (L) (L) 2007 No Flow No Flow 2008 2009 2010 131 16 116 18 68 18 71 17 2011 295 14 268 15  Total Outflow (L)  Total Outflow (mm)  36 31  18 15  297 194 203 202  83 54 97 97  66 22 23 38  18 6 11 18  147 134 84 88  41 37 40 42  309 86 283 79 146 24/Oct/2011 132 14 70 (86*) TIII West 5/Jun/2011 (180*) 5/Jun/2011 20/Oct/2011 170 15 185 88 TIII East *Values in parentheses include the estimated outflow missed in 2011 at the Type III West active-zone lysimeter TI West TI East  267  A.7  FAO-56 Penman-Monteith evaporation and net infiltration analysis Figures A-155 to A-161 present the FAO-56 Penman-Monteith results from the  analysis described in Chapter 3. Reference evaporation, rainfall, actual evaporation, and net infiltration are plotted for the 2007 through 2011 analyses. Table A-11 summarizes the results from all years.  268  Figure ‎A-155  2007 FAO-PM method calculation, using rainfall at the crown of the Type I test pile  Figure ‎A-156  2008 FAO-PM method calculation, using rainfall at the crown of the Type III test pile  269  Figure ‎A-157  2008 FAO-PM method calculation, using rainfall at the Type I active-zone lysimeters  Figure ‎A-158  2008 FAO-PM method calculation, using rainfall at the Type III active-zone lysimeters  270  Figure ‎A-159  2009 FAO-PM method calculation, using rainfall representative of all areas in the Test  Piles Research Area  Figure ‎A-160  2010 FAO-PM method calculation, using rainfall representative of all areas in the Test  Piles Research Area  271  Figure ‎A-161  2011 FAO-PM method calculation, using rainfall representative of all areas in the Test  Piles Research Area  272  Table ‎A-11  FAO-PM method results  Year 2007 (Type I Test Pile) 2007 (Type III Test Pile) 2008 (Type I AZLs) 2008 (Type III AZLs) 2009 2010 2011  92  Reference evaporation (mm) 342  Actual evaporation (mm) 48  Net infiltration (deep percolation) (mm) 40  153  342  58  92  154 180 74 98 146  320 320 295 357 335  72 76 61 54 59  83 105 10 43 83  Rainfall (mm)  273  Appendix B Tipping buckets  B.1  Design The main basal drain tipping buckets (used to monitor outflows during this thesis)  were fabricated at the University of British Columbia, using a design described by Neuner (2009) and modified by Momeyer (in progress). Figures B-1 and B-2 are AutoCad drawings of the bucket and box design. The buckets and boxes of each tipping bucket were fabricated from acrylic. Stainless steel was used for the axles running through each bucket, and rare earth magnets and Texas Electronics reed switches were used as a means to send a pulse to the data logger after each tip of the bucket. Figures B-3 to B-5 are pictures of improvements devised by Momeyer (in progress) that are not included in the AutoCad design drawings. The original basal drain tipping buckets used in 2007 were fabricated by Plasticsmith, Inc. (Vancouver, B.C., Canada). The 1Bxxdrn13 (Type I basal drain), 3BNxdrn15 (Type III North basal drain), and 3BSxdrn15 (Type III South basal drain) tipping buckets were replaced with UBC fabricated tipping buckets in 2008. The 3BNxdrn15 and 3BSxdrn15 tipping buckets were replaced again in 2010, while the 1Bxxdrn13 tipping bucket was replaced in 2011. These new, 2010 and 2011 tipping buckets (Figures B-1 and B-2) had increased capacity to allow for accurate measurement of high flow rates.  274  Figure ‎B-1  AutoCad drawing of the tipping bucket box design. Original design by Neuner (2009), updated by Momeyer (in progress).  275  Figure ‎B-2  AutoCad drawing of the tipping bucket design. Original design by Neuner (2009),  updated by Momeyer (in progress).  276  Figure ‎B-3  Support triangles reinforcing the connections between the side-walls and base of the  tipping bucket box. Source: Modified from Momeyer (in progress).  Figure ‎B-4  Bumpers to cushion the bucket during tips and level adjusters to facilitate easy levelling  of the bucket. Source: Modified from Momeyer (in progress).  277  Figure ‎B-5  Non-threaded drainage system at the base of the tipping bucket box. Source: Modified  from Momeyer (in progress).  278  B.2  Calibration Tipping buckets were calibrated to determine the relationship between tip-time and  flow-rate. Figure B-6 is an example of a calibration curve. Table B-1 lists the calibration equations for the tipping buckets in the Test Piles Research Area from 2008 through 2011.  279  Figure ‎B-6  Calibration curve for the 1Bxxdrn13 (Type I test pile basal drain) tipping bucket.  Calibration was conducted on July 19th, 2011.  Table ‎B-1  Tipping bucket calibration equations for 2008 through 2011  Location 1Bxxdrn13 3BNxdrn15 3BSxdrn15 CBxxdrn13 AZLs and BCLs Young Model 2202 tipping buckets  Date Range 2008 to July 19th, 2011 at 15:00 July 19th, 2011 at 15:00 through 2011 2008 through June 19th, 2010 June 20th, 2010 through 2011 2008 through June 16th,2010 June 17th, 2010 through 2011 2008 through 2011  Calibration equation* y = 828.46x-1.052 y = 664.16x-1.035 y = 223.54x-1.0421 y = 586.01x-1.0329 y = 203.41x-1.0792 y = 593.8x-1.0297 y = 199.82x-1.074  2008 through 2011  y = 2.6425x-1.1124  *Where: y = flow rate (mL/s) and x = time between tips (s)  280  Appendix C Field-permeameter Section 2.4.1.2 reported values for hydraulic conductivity and porosity of bulk wasterock in a 32 m3 field-permeameter. More detailed results from the draindown and constant head tests are given here. Construction and experimental set-up details were described in Section 2.2.3 and Figures 2-9 and 2-10.  C.1  Draindown test A draindown test was conducted at the field-permeameter to determine the fraction of  bulk-porosity attributed to matrix-pores and macro-pores. Bulk porosity was calculated by measuring the volume of water required to saturate the field-permeameter from the base up, taking into account initial moisture content of the matrix material and the volume of water required to saturate the crush material at the base. Momeyer (in progress) was able to measure the initial moisture content of the matrix material during construction of the fieldpermeameter and calculated a bulk-porosity of 0.27. Initial moisture contents could not be measured‎for‎the‎draindown‎test‎described‎here‎and,‎thus,‎Momeyer’s‎bulk-porosity estimate of 0.27 was adopted. Atmospheric conditions were induced at the base of the fieldpermeameter by opening the outflow drain at the base. Outflow rate and volume, during draindown, were monitored manually using a stop watch and a 20 L bucket. Watertable elevation within the field-permeameter, during draindown, was monitored in the manometers using a water-level tape. It was assumed that the volume of outflow recorded from fullysaturated conditions to the onset of unsaturated conditions at the base of the fieldpermeameter approximates the volume of pore-space in the macropores. Conversely, the volume of water remaining at the onset of unsaturated conditions at the base of the field-  281  permeameter appriximates the volume of pore-space in the matrix fraction. This is due to the finer, <5 mm diameter, matrix fraction exhibiting capillarity and, thus, transporting and storing water under tension (Yazdani et al., 2000). Figure C-1 shows flow rate over time at the outflow point (secondary y-axis) and the water level within the permeameter (primary yaxis) during the draindown test. The onset of unsaturated conditions at the base of the fieldpermeameter is followed by a short period of gravel crush drainage and then a sharp drop in flow rate as the entirety of the material above the screened-drain becomes unsaturated. Figure C-2 presents the volume of water recovered during draindown as a percent of the total initial volume of water taken to saturate the field-permeameter (secondary y-axis) and water level within the field-permeameter during the draindown test (primary y-axis). It was calculated that 70 % of the bulk-porosity was attributable to macropore volume and 30 % of the bulkporosity was attributable to matrix-pore volume. This corresponds to 19 % and 8 % macroporosity and matrix porosity, respectively. Uncertainty in porosity values arises because outflow during the draindown test was restricted through ½-inch tubing to allow accurate outflow volumes to be measured. By restricting the outflow, water in the unsaturated zone (that develops as the water table drops during draindown) may migrate to the water table. This volume of water would then be interpreted as originating from macropores. Therefore, the estimate of matrix-porosity represents a minimum estimate.  C.2  Constant head tests A series of constant head tests were conducted at the field-permeameter to determine  the saturated hydraulic conductivity (Ksat) of the bulk waste-rock. Constant head conditions were achieved by using a constant head reservoir to control the flow rate into the base of the  282  permeameter. Hydraulic gradient across the permeameter during the tests was averaged from the three manometers screened at the base of the field-permeameter. Discharge rate was measured at the outflow port at the top of the field-permeameter several times during a single test and averaged. The field-permeameter was allowed to come to equilibrium during each test. Table C-1 summarizes the seven constant head tests conducted at the field-permeameter. The average Ksat was calculated to be 6.0x10-3 m/s for the bulk waste-rock.  283  Figure ‎C-1  Field-permeameter draindown test showing flow rate measured at the base of the field-permeameter and watertable level within the  field-permeameter  284  Figure ‎C-2  Field-permeameter draindown test showing volume of outflow as a percent of the total initial volume of water taken to saturate the  field-permeameter and watertable level within the field-permeameter  285  Table ‎C-1  Trial # 1 2 3 4 5 6 7 Average  Constant head test results  Discharge (Q) (m3/s) 3.8x10-4 3.3x10-4 2.9x10-4 4.7x10-4 2.9x10-4 5.3x10-4 3.7x10-4  Gradient (dh/dl) (m/m) 0.0022 0.0039 0.0028 0.0033 0.0050 0.0056 0.0056  Ksat (m/s) 1.2x10-2 5.4x10-3 6.1x10-3 9.0x10-3 3.4x10-3 5.9x10-3 4.1x10-3 6.0x10-3  286  Appendix D MATLAB analysis scripts The programming language MATLAB was used to process and plot a majority of the hydrology dataset. Raw TDR, tensiometer, and tipping bucket data were compiled as text files and then run through MATLAB scripts to transform the data into the required format. The following sub-sections provide examples of scripts created to process each type of data. MATLAB scripts have been modified from work by Neuner (2009) and Momeyer (in progress).  287  D.1  Time domain reflectrometry analysis example  %Script name: getRawTDR_TIIIsmooth %Author: Modified by Nathan Fretz from Matthew Neuner and Steven Momeyer %Project: Diavik Waste-Rock Project %Plots new table based data from two raw datalogger files. %Employs moving average filter to average vmc over every 6 hrs. %% function [TDR_51,TDR_52,TDR_53,TDR_54,TDR_55,TDR_56,TDR_57,TDR_58,TDR_59,... TDR_60,TDR_61,TDR_62,TDR_63,TDR51,TDR52,TDR53,TDR54,TDR55,TDR56,... TDR57,TDR58,TDR59,TDR60,TDR61,TDR62,TDR63,TDR51smooth,TDR52smooth,... TDR53smooth,TDR54smooth,TDR55smooth,TDR56smooth,TDR57smooth,... TDR58smooth,TDR59smooth,TDR60smooth,TDR61smooth,TDR62smooth... ,TDR63smooth] = getRawTDR_TIIIsmooth(inName,inName2) %inName: Write directory of file with TDR 1to7 here. %inName2: Write directory of file with TDR 8to13 here. %Create data file containing calibration data for TDR Cal1 = [1:50;1:50;1:50]'; %Filler array to match ArrayID with Row number %ArrayID, Toff, Tair T3_Calibration = [51,1.21460071799812, 2.38739928200188; 52,1.15110071799812, 2.38739928200188; 53,1.06676803571631, 2.64123196428369; 54,1.2654514610408, 2.40111996753063; 55,1.19616574675509, 2.40111996753063; 56,1.17688003246937, 2.40111996753063; 57,0.839443922543831, 2.72355607745617; 58,0.797443922543831, 2.72355607745617; 59,0.811343922543831, 2.72355607745617; 60,1.13485763311876, 2.44914236688124; 61,1.51000414564185, 2.31879585435815; 62,1.62124380287748,2.32565619712252; 63,1.19539294596654,2.34280705403346]; Cal2 = [64:79;64:79;64:79]'; %Filler array to match ArrayID with Row number T1_Calibration = [80,1.09722,2.3300; 81,81,81; 82,82,82; 83,0.98350,2.5900; 84,84,84; 85,85,85; 86,1.21381,2.32333; 87,1.13314,2.32000; 88,88,88; 89,89,89]; CT_Calibration = [90,1.368974,2.268226; 91,1.578055,2.254845; 92,1.56091,2.25819; 93,1.622819,2.264881;  288  94,1.547283,2.274917; 95,1.421601,2.251499; 96,1.453955,2.254845; 97,1.285855,2.25845; 98,1.328692,2.281608; 99,1.408465,2.261535; 100,1.205501,2.251499]; TDR_Calibration = vertcat(Cal1,T3_Calibration,Cal2,T1_Calibration,... CT_Calibration); %Temperatures used in basic processing CalTemp = 22; MeanTemp = 5; %File format constants %timestamp, rec id, arr_id, min, sec, avg DDtt, std dateFormatString = '"%f-%f-%f %f:%f:%f", %f, %f, %f, %f, %f, %f\n'; nColumns = 12; secondsColumn = 6; minuteColumn = 5; hourColumn = 4; daysColumn = 3; monthsColumn = 2; yearsColumn = 1; %recordNumberColumn = 7; ArrayIDColumn = 8; DDttColumn = 11; %stdDev = 12;  %Read data from file Data = txt2mat(inName,4,nColumns,dateFormatString); Data2 = txt2mat(inName2,4,nColumns,dateFormatString); Data = vertcat(Data, Data2); %Calculate the time in days using matlab's datenum dateVec = [Data(:,yearsColumn) Data(:,monthsColumn) Data(:,daysColumn)... Data(:, hourColumn) Data(:, minuteColumn) Data(:,secondsColumn)]; tRaw = datenum(dateVec); %Add column for sorting tRaw(:,2) = [Data(:,ArrayIDColumn)]; %This allows file to contain multiple %TDR array IDs. They will all be grouped and then need to be separated. %Add raw travel time into array tRaw(:,3) = [Data(:,DDttColumn)]; %Sort arrays into separate array %row counters, j overall counter to step through each row %count51..., counter to place values next to each other in new array j = 1; count51 = 1; count52 = 1;  289  count53 count54 count55 count56 count57 count58 count59 count60 count61 count62 count63  = = = = = = = = = = =  1; 1; 1; 1; 1; 1; 1; 1; 1; 1; 1;  %pre-allocate for speed L = 5000; TDR_51 = ones(L,3); TDR_52 = ones(L,3); TDR_53 = ones(L,3); TDR_54 = ones(L,3); TDR_55 = ones(L,3); TDR_56 = ones(L,3); TDR_57 = ones(L,3); TDR_58 = ones(L,3); TDR_59 = ones(L,3); TDR_60 = ones(L,3); TDR_61 = ones(L,3); TDR_62 = ones(L,3); TDR_63 = ones(L,3); maxsize = size(tRaw,1); while j < maxsize+1; if tRaw(j,2) == 51 && tRaw(j,3) > 5 && tRaw(j,3) < 10; TDR_51(count51,:) = [tRaw(j,1),tRaw(j,2), tRaw(j,3)]; count51 = count51+1; elseif tRaw(j,2) == 52 && tRaw(j,3) > 5 && tRaw(j,3) < TDR_52(count52,:) = [tRaw(j,1),tRaw(j,2), tRaw(j,3)]; count52 = count52+1; elseif tRaw(j,2) == 53 && tRaw(j,3) > 5 && tRaw(j,3) < TDR_53(count53,:) =[tRaw(j,1),tRaw(j,2), tRaw(j,3)]; count53 = count53+1; elseif tRaw(j,2) == 54 && tRaw(j,3) > 5 && tRaw(j,3) < TDR_54(count54,:) = [tRaw(j,1),tRaw(j,2), tRaw(j,3)]; count54 = count54+1; elseif tRaw(j,2) == 55 && tRaw(j,3) > 5 && tRaw(j,3) < TDR_55(count55,:) = [tRaw(j,1),tRaw(j,2), tRaw(j,3)]; count55 = count55+1; elseif tRaw(j,2) == 56 && tRaw(j,3) > 5 && tRaw(j,3) < TDR_56(count56,:) = [tRaw(j,1),tRaw(j,2), tRaw(j,3)]; count56 = count56+1; elseif tRaw(j,2) == 57 && tRaw(j,3) > 5 && tRaw(j,3) < TDR_57(count57,:) = [tRaw(j,1),tRaw(j,2), tRaw(j,3)]; count57 = count57+1; elseif tRaw(j,2) == 58 && tRaw(j,3) > 5 && tRaw(j,3) < TDR_58(count58,:) = [tRaw(j,1),tRaw(j,2), tRaw(j,3)]; count58 = count58+1; elseif tRaw(j,2) == 59 && tRaw(j,3) > 5 && tRaw(j,3) < TDR_59(count59,:) = [tRaw(j,1),tRaw(j,2), tRaw(j,3)];  10; 10; 10; 10; 10; 10; 10; 10;  290  count59 = count59+1; elseif tRaw(j,2) == 60 && tRaw(j,3) > 5 && tRaw(j,3) < TDR_60(count60,:) = [tRaw(j,1),tRaw(j,2), tRaw(j,3)]; count60 = count60+1; elseif tRaw(j,2) == 61 && tRaw(j,3) > 5 && tRaw(j,3) < TDR_61(count61,:) = [tRaw(j,1),tRaw(j,2), tRaw(j,3)]; count61 = count61+1; elseif tRaw(j,2) == 62 && tRaw(j,3) > 5 && tRaw(j,3) < TDR_62(count62,:) = [tRaw(j,1),tRaw(j,2), tRaw(j,3)]; count62 = count62+1; elseif tRaw(j,2) == 63 && tRaw(j,3) > 5 && tRaw(j,3) < TDR_63(count63,:) = [tRaw(j,1),tRaw(j,2), tRaw(j,3)]; count63 = count63+1; end; j = j+1; end;  10; 10; 10; 10;  %Convert DDtt to T/Tair: DDtt - Toff/Tair TDR_51(:,4) = (TDR_51(:,3)-TDR_Calibration(51,2))./TDR_Calibration(51,3); %Convert to T/Tair corrected: T/Tair-(CalTemp %MeanTemp)*((-0.006*T/Tair)+(2.645*0.006)) TDR_51(:,5) = TDR_51(:,4)-(CalTemp - MeanTemp).*((-0.006*TDR_51(:,4))+... (2.645*0.006)); %Convert to VMC: 0.093(T/Tair)^4 - 0.983(T/Tair)^3 + 3.74(T/Tair)^2 %5.90(T/Tair) + 3.29 TDR_51(:,6) = (0.093.*TDR_51(:,5).^4 - 0.983.*TDR_51(:,5).^3 +... 3.74.*TDR_51(:,5).^2 - 5.90.*TDR_51(:,5) + 3.29).*100; %Now do the above 3 steps for the rest of the probes. TDR_52(:,4) = (TDR_52(:,3)-TDR_Calibration(52,2))./TDR_Calibration(52,3); TDR_52(:,5) = TDR_52(:,4)-(CalTemp - MeanTemp).*((-0.006*TDR_52(:,4))+... (2.645*0.006)); TDR_52(:,6) = (0.093.*TDR_52(:,5).^4 - 0.983.*TDR_52(:,5).^3 +... 3.74.*TDR_52(:,5).^2 - 5.90.*TDR_52(:,5) + 3.29).*100; TDR_53(:,4) = (TDR_53(:,3)-TDR_Calibration(53,2))./TDR_Calibration(53,3); TDR_53(:,5) = TDR_53(:,4)-(CalTemp - MeanTemp).*((-0.006*TDR_53(:,4))+... (2.645*0.006)); TDR_53(:,6) = (0.093.*TDR_53(:,5).^4 - 0.983.*TDR_53(:,5).^3 +... 3.74.*TDR_53(:,5).^2 - 5.90.*TDR_53(:,5) + 3.29).*100; TDR_54(:,4) = (TDR_54(:,3)-TDR_Calibration(54,2))./TDR_Calibration(54,3); TDR_54(:,5) = TDR_54(:,4)-(CalTemp - MeanTemp).*((-0.006*TDR_54(:,4))+... (2.645*0.006)); TDR_54(:,6) = (0.093.*TDR_54(:,5).^4 - 0.983.*TDR_54(:,5).^3 +... 3.74.*TDR_54(:,5).^2 - 5.90.*TDR_54(:,5) + 3.29).*100; TDR_55(:,4) = (TDR_55(:,3)-TDR_Calibration(55,2))./TDR_Calibration(55,3); TDR_55(:,5) = TDR_55(:,4)-(CalTemp - MeanTemp).*((-0.006*TDR_55(:,4))+... (2.645*0.006)); TDR_55(:,6) = (0.093.*TDR_55(:,5).^4 - 0.983.*TDR_55(:,5).^3 +... 3.74.*TDR_55(:,5).^2 - 5.90.*TDR_55(:,5) + 3.29).*100;  291  TDR_56(:,4) = (TDR_56(:,3)-TDR_Calibration(56,2))./TDR_Calibration(56,3); TDR_56(:,5) = TDR_56(:,4)-(CalTemp - MeanTemp).*((-0.006*TDR_56(:,4))+... (2.645*0.006)); TDR_56(:,6) = (0.093.*TDR_56(:,5).^4 - 0.983.*TDR_56(:,5).^3 +... 3.74.*TDR_56(:,5).^2 - 5.90.*TDR_56(:,5) + 3.29).*100; TDR_57(:,4) = (TDR_57(:,3)-TDR_Calibration(57,2))./TDR_Calibration(57,3); TDR_57(:,5) = TDR_57(:,4)-(CalTemp - MeanTemp).*((-0.006*TDR_57(:,4))+... (2.645*0.006)); TDR_57(:,6) = (0.093.*TDR_57(:,5).^4 - 0.983.*TDR_57(:,5).^3 +... 3.74.*TDR_57(:,5).^2 - 5.90.*TDR_57(:,5) + 3.29).*100; TDR_58(:,4) = (TDR_58(:,3)-TDR_Calibration(58,2))./TDR_Calibration(58,3); TDR_58(:,5) = TDR_58(:,4)-(CalTemp - MeanTemp).*((-0.006*TDR_58(:,4))+... (2.645*0.006)); TDR_58(:,6) = (0.093.*TDR_58(:,5).^4 - 0.983.*TDR_58(:,5).^3 +... 3.74.*TDR_58(:,5).^2 - 5.90.*TDR_58(:,5) + 3.29).*100; TDR_59(:,4) = (TDR_59(:,3)-TDR_Calibration(59,2))./TDR_Calibration(59,3); TDR_59(:,5) = TDR_59(:,4)-(CalTemp - MeanTemp).*((-0.006*TDR_59(:,4))+... (2.645*0.006)); TDR_59(:,6) = (0.093.*TDR_59(:,5).^4 - 0.983.*TDR_59(:,5).^3 +... 3.74.*TDR_59(:,5).^2 - 5.90.*TDR_59(:,5) + 3.29).*100; TDR_60(:,4) = (TDR_60(:,3)-TDR_Calibration(60,2))./TDR_Calibration(60,3); TDR_60(:,5) = TDR_60(:,4)-(CalTemp - MeanTemp).*((-0.006*TDR_60(:,4))+... (2.645*0.006)); TDR_60(:,6) = (0.093.*TDR_60(:,5).^4 - 0.983.*TDR_60(:,5).^3 +... 3.74.*TDR_60(:,5).^2 - 5.90.*TDR_60(:,5) + 3.29).*100; TDR_61(:,4) = (TDR_61(:,3)-TDR_Calibration(61,2))./TDR_Calibration(61,3); TDR_61(:,5) = TDR_61(:,4)-(CalTemp - MeanTemp).*((-0.006*TDR_61(:,4))+... (2.645*0.006)); TDR_61(:,6) = (0.093.*TDR_61(:,5).^4 - 0.983.*TDR_61(:,5).^3 +... 3.74.*TDR_61(:,5).^2 - 5.90.*TDR_61(:,5) + 3.29).*100; TDR_62(:,4) = (TDR_62(:,3)-TDR_Calibration(62,2))./TDR_Calibration(62,3); TDR_62(:,5) = TDR_62(:,4)-(CalTemp - MeanTemp).*((-0.006*TDR_62(:,4))+... (2.645*0.006)); TDR_62(:,6) = (0.093.*TDR_62(:,5).^4 - 0.983.*TDR_62(:,5).^3 +... 3.74.*TDR_62(:,5).^2 - 5.90.*TDR_62(:,5) + 3.29).*100; TDR_63(:,4) = (TDR_63(:,3)-TDR_Calibration(63,2))./TDR_Calibration(63,3); TDR_63(:,5) = TDR_63(:,4)-(CalTemp - MeanTemp).*((-0.006*TDR_63(:,4))+... (2.645*0.006)); TDR_63(:,6) = (0.093.*TDR_63(:,5).^4 - 0.983.*TDR_63(:,5).^3 +... 3.74.*TDR_63(:,5).^2 - 5.90.*TDR_63(:,5) + 3.29).*100; %TDR_51 %TDR_52 %TDR_53 %TDR_54 %TDR_55 %TDR_56 %TDR_57  = = = = = = =  31N2tdr01 31N2tdr07 31N2tdr09 31S2tdr01 31S2tdr03 31S2tdr05 31S2tdr09  292  %TDR_58 %TDR_59 %TDR_60 %TDR_61 %TDR_62 %TDR_63  = = = = = =  32N2tdr07 33N2tdr01 33N2tdr03 33N2tdr05 33S2tdr02 33S2tdr03  %Create TDR51 = TDR52 = TDR53 = TDR54 = TDR55 = TDR56 = TDR57 = TDR58 = TDR59 = TDR60 = TDR61 = TDR62 = TDR63 =  new table with Date/Time in column 1 and vmc in column 2. [TDR_51(:,1),TDR_51(:,6)]; [TDR_52(:,1),TDR_52(:,6)]; [TDR_53(:,1),TDR_53(:,6)]; [TDR_54(:,1),TDR_54(:,6)]; [TDR_55(:,1),TDR_55(:,6)]; [TDR_56(:,1),TDR_56(:,6)]; [TDR_57(:,1),TDR_57(:,6)]; [TDR_58(:,1),TDR_58(:,6)]; [TDR_59(:,1),TDR_59(:,6)]; [TDR_60(:,1),TDR_60(:,6)]; [TDR_61(:,1),TDR_61(:,6)]; [TDR_62(:,1),TDR_62(:,6)]; [TDR_63(:,1),TDR_63(:,6)];  %Now filter data using a moving average filter. %Average every 12 data points. This averages over 6 hours. span = 12; %Size of the averaging window. window = ones(span,1)/span; TDR51smooth(:,1) = TDR51(:,1); TDR51smooth(:,2) = convn(TDR51(:,2),window,'same'); TDR52smooth(:,1) = TDR52(:,1); TDR52smooth(:,2) = convn(TDR52(:,2),window,'same'); TDR53smooth(:,1) = TDR53(:,1); TDR53smooth(:,2) = convn(TDR53(:,2),window,'same'); TDR54smooth(:,1) = TDR54(:,1); TDR54smooth(:,2) = convn(TDR54(:,2),window,'same'); TDR55smooth(:,1) = TDR55(:,1); TDR55smooth(:,2) = convn(TDR55(:,2),window,'same'); TDR56smooth(:,1) = TDR56(:,1); TDR56smooth(:,2) = convn(TDR56(:,2),window,'same'); TDR57smooth(:,1) = TDR57(:,1); TDR57smooth(:,2) = convn(TDR57(:,2),window,'same'); TDR58smooth(:,1) = TDR58(:,1); TDR58smooth(:,2) = convn(TDR58(:,2),window,'same'); TDR59smooth(:,1) = TDR59(:,1); TDR59smooth(:,2) = convn(TDR59(:,2),window,'same');  293  TDR60smooth(:,1) = TDR60(:,1); TDR60smooth(:,2) = convn(TDR60(:,2),window,'same'); TDR61smooth(:,1) = TDR61(:,1); TDR61smooth(:,2) = convn(TDR61(:,2),window,'same'); TDR62smooth(:,1) = TDR62(:,1); TDR62smooth(:,2) = convn(TDR62(:,2),window,'same'); TDR63smooth(:,1) = TDR63(:,1); TDR63smooth(:,2) = convn(TDR63(:,2),window,'same');  294  D.2  Tensiometer example  %Script name: getRawTensiometer_TIII %Author: Modified by Nathan Fretz from Matthew Neuner and Steven Momeyer %Project: Diavik Waste-Rock Project %This script opens raw data files and creates an array %containing processed data from the tensiometers at the Type III test pile. %% function [Tensiometers_TIII] = getRawTensiometer_TIII(inName) %inName: Write directory of file here. %'Tensiometers_TIII' is an array with 1)Date 2)raw tension of tensiometer_4 %3)Pressure head in head space of tensiometer_1 4)Pressure head at cup of %tensiometer_4 5)Hydraulic head at cup of tensiometer_4 %6)-17) follow same pattern as 2)3)4)5) for tensiometer_5, tensiometer_6 %and tensiometer_7. %AZL tensiometers: %Tension_4 %Tension_5 %Tension_6 %Tension_7 %First variable in matrix is 'height of water above porous cup. %Second variable in matrix is 'elevation of porous cup' (Ground surface is %reference elevation at 0m. elevation_4 = [0.76,-0.60]; elevation_5 = [1.37,-1.20]; elevation_6 = [0.76,-0.60]; elevation_7 = [1.37,-1.20]; %Specific weight of water (constant value used) Sw_H2O = 9.8057; %File format constants %"TIMESTAMP","RECORD","Arr_ID3","Minute","seconds","Tension_8","Tension_9" , %"Tension10","Tension11" dateFormatString = '"%f-%f-%f %f:%f:%f", %f, %f, %f, %f, %f, %f, %f, %f\n'; nColumns = 14; minuteColumn = 5; hourColumn = 4; secondsColumn = 6; daysColumn = 3; monthsColumn = 2; yearsColumn = 1; %recordNumberColumn = 7; %arrayID = 8; Tension_4 = 11; %3Ta  295  Tension_5 = 12; Tension_6 = 13; Tension_7 = 14; Data = txt2mat(inName,4,nColumns,dateFormatString); %Calculate the time in days using matlab's datenum dateVec = [Data(:,yearsColumn) Data(:,monthsColumn) Data(:,daysColumn)... Data(:, hourColumn) Data(:, minuteColumn) Data(:,secondsColumn)]; tRaw = datenum(dateVec); %Place raw tension readings (from datalogger into array). %The datalogger program converts voltage to soil tension. TensionData = [Data(:,Tension_4),Data(:,Tension_5),Data(:,Tension_6),... Data(:,Tension_7)]; %preallocate piezo array for speed L = size(TensionData,1); Tensiometers_TIII(1:L,17) = 1; %Tensiometer_4 %Date: Tensiometers_TIII(:,1) = tRaw(:,1); %raw tension value (kPa): Tensiometers_TIII(:,2) = TensionData(:,1).*-1; %Pressure head in head space: Tensiometers_TIII(:,3) = Tensiometers_TIII(:,2)./Sw_H2O; %Pressure head at cup: Tensiometers_TIII(:,4) = Tensiometers_TIII(:,3)+elevation_4(1,1); %Hydraulic head at cup: Tensiometers_TIII(:,5) = Tensiometers_TIII(:,4)+elevation_4(1,2); %Tensiometer_5 Tensiometers_TIII(:,6) Tensiometers_TIII(:,7) Tensiometers_TIII(:,8) Tensiometers_TIII(:,9)  = = = =  TensionData(:,2).*-1; Tensiometers_TIII(:,6)./Sw_H2O; Tensiometers_TIII(:,7)+elevation_5(1,1); Tensiometers_TIII(:,8)+elevation_5(1,2);  %Tensiometer_6 Tensiometers_TIII(:,10) Tensiometers_TIII(:,11) Tensiometers_TIII(:,12) Tensiometers_TIII(:,13)  = = = =  TensionData(:,3).*-1; Tensiometers_TIII(:,10)./Sw_H2O; Tensiometers_TIII(:,11)+elevation_6(1,1); Tensiometers_TIII(:,12)+elevation_6(1,2);  %Tensiometer_7 Tensiometers_TIII(:,14) Tensiometers_TIII(:,15) Tensiometers_TIII(:,16) Tensiometers_TIII(:,17)  = = = =  TensionData(:,4).*-1; Tensiometers_TIII(:,14)./Sw_H2O; Tensiometers_TIII(:,15)+elevation_7(1,1); Tensiometers_TIII(:,16)+elevation_7(1,2);  296  D.3  Tipping bucket example This first script sorts the raw tipping bucket data (the Type III basal drain and basal  collection lysimeter tipping buckets in this example) into individual files, each containing a single tipping bucket. %Script name: getRawTips3Ba %Author: Modified by Nathan Fretz from Matthew Neuner and Steven Momeyer %Project: Diavik Waste-Rock Project %% function [T1_drain,tb_1,tb_2,tb_3,tb_4,tb_5,tb_6,tb_7,tb_8,tb_9,tb_10,... tb_11,tb_12,tb_13] = getRawTips3Ba(inName) %inName: Write directory of file here. %This script opens raw data files from the datalogger and sorts the tipping %buckets based on ArrayID %Creates an array for each tipping bucket with Date in column 1 and time %between tips (in days) in column 2. %tb_1 = 3BNBlys2W & 3BNBlys2E %tb_2 = 3BNBlys4W %tb_3 = 3BNBlys4E %tb_4 = 3BNClys2W & 3BNClys2E %tb_5 = 3BNClys4W %tb_6 = 3BNClys4E %tb_7 = 3BSClys2W & 3BSClys2E %tb_8 = 3BSClys4W %tb_9 = 3BSClys4E %tb_10 = 1BWBlys Cluster %tb_11 = 1BWClys Cluster %tb_12 = 1BEClys Cluster %tb_13 = 1Bxxdrn13 %File format constants. %"TIMESTAMP","RECORD","Arr_ID","Minutes","Second" dateFormatString = '"%f-%f-%f %f:%f:%f",%f,%f,%f,%f\n'; nColumns = 10; minuteColumn = 5; hourColumn = 4; secondsColumn = 6; daysColumn = 3; monthsColumn = 2; yearsColumn = 1; %recordNumberColumn = 7; arrayID = 8; %Filter for taking derivatives. diffFilter = [1,-1];  297  %Read data from file. %Data = txt2mat(file location, header rows, columns, format string); Data = txt2mat(inName,4,nColumns,dateFormatString); %Calculate the time in days using matlab's datenum. dateVec = [Data(:,yearsColumn) Data(:,monthsColumn) Data(:,daysColumn)... Data(:, hourColumn) Data(:, minuteColumn) Data(:,secondsColumn)]; tRaw = datenum(dateVec); %Add the array ID to be used for sorted. tRaw(:,2) = Data(:,arrayID); %Outputs all data before sorting and creating separate arrays for each ID. T1_drain = [tRaw(:,1), tRaw(:,2)]; %Separate and sort array IDs (basal drains and basal lysimeters). %j is the overall counter to step through each row. %count1 etc... are counters to place values next to each other in new %array j = 1; count1 = 1; count2 = 1; count3 = 1; count4 = 1; count5 = 1; count6 = 1; count7 = 1; count8 = 1; count9 = 1; count10 = 1; count11 = 1; count12 = 1; count13 = 1; %Preallocate new arrays for speed. tip_1 = zeros(1000000,2); tip_2 = zeros(1000000,2); tip_3 = zeros(1000000,2); tip_4 = zeros(1000000,2); tip_5 = zeros(1000000,2); tip_6 = zeros(1000000,2); tip_7 = zeros(1000000,2); tip_8 = zeros(1000000,2); tip_9 = zeros(1000000,2); tip_10 = zeros(1000000,2); tip_11 = zeros(1000000,2); tip_12 = zeros(1000000,2); tip_13 = zeros(1000000,2); %Dummy arrays to solve errors which were occur later in the %program if arrays are called that have not been created because %there was no data recorded at the logger. tb_1(1,2)= -1; tb_2(1,2)= -1; tb_3(1,2)= -1; tb_4(1,2)= -1;  298  tb_5(1,2)= -1; tb_6(1,2)= -1; tb_7(1,2)= -1; tb_8(1,2)= -1; tb_9(1,2)= -1; tb_10(1,2)= -1; tb_11(1,2)= -1; tb_12(1,2)= -1; tb_13(1,2)= -1; %Scans through each row, checking the array ID of column 2 and places the %time and array ID in its own array. maxsize = size(T1_drain,1); while j < maxsize+1; if T1_drain(j,2) == 1; %arrayID column tip_1(count1,:) = [T1_drain(j,1),T1_drain(j,2)]; count1 = count1+1; elseif T1_drain(j,2) == 2; tip_2(count2,:) = [T1_drain(j,1),T1_drain(j,2)]; count2 = count2+1; elseif T1_drain(j,2) == 3; tip_3(count4,:) = [T1_drain(j,1),T1_drain(j,2)]; count3 = count3+1; elseif T1_drain(j,2) == 4; tip_4(count4,:) = [T1_drain(j,1),T1_drain(j,2)]; count4 = count4+1; elseif T1_drain(j,2) == 5; tip_5(count5,:) = [T1_drain(j,1),T1_drain(j,2)]; count5 = count5+1; elseif T1_drain(j,2) == 6; tip_6(count6,:) = [T1_drain(j,1),T1_drain(j,2)]; count6 = count6+1; elseif T1_drain(j,2) == 7; tip_7(count7,:) = [T1_drain(j,1),T1_drain(j,2)]; count7 = count7+1; elseif T1_drain(j,2) == 8; tip_8(count8,:) = [T1_drain(j,1),T1_drain(j,2)]; count8 = count8+1; elseif T1_drain(j,2) == 9; tip_9(count9,:) = [T1_drain(j,1),T1_drain(j,2)]; count9 = count9+1; elseif T1_drain(j,2) == 10; tip_10(count10,:) = [T1_drain(j,1),T1_drain(j,2)]; count10 = count10+1; elseif T1_drain(j,2) == 11; tip_11(count11,:) = [T1_drain(j,1),T1_drain(j,2)]; count11 = count11+1; elseif T1_drain(j,2) == 12; tip_12(count12,:) = [T1_drain(j,1),T1_drain(j,2)]; count12 = count12+1; elseif T1_drain(j,2) == 13; tip_13(count13,:) = [T1_drain(j,1),T1_drain(j,2)]; count13 = count13+1; end; j = j+1; end;  299  %Calculate differences in time in days in each array and creates new array %with time stamp and tip day (in days) if tip_1(1,2) == 1 tip_1(:,2) = []; %Removes arrayID column to correctly call convSame dtRaw_1 = convSame(diffFilter, tip_1(:)); dtRaw_1 = dtRaw_1(2:end); tb_1 = [tip_1(2:end,:), dtRaw_1]; [r1,c1] = find(tip_1==0); %Finds first zero from preallocation to if r1 > 1; %trim final array. tb_1 = tb_1(1:r1-1,:); end end if tip_2(1,2) == 2 tip_2(:,2) = []; dtRaw_2 = convSame(diffFilter, tip_2(:)); dtRaw_2 = dtRaw_2(2:end); tb_2 = [tip_2(2:end,:), dtRaw_2]; [r2,c1] = find(tip_2==0); if r2 > 1; tb_2 = tb_2(1:r2-1,:); end end if tip_3(1,2) == 3 tip_3(:,2) = []; dtRaw_3 = convSame(diffFilter, tip_3(:)); dtRaw_3 = dtRaw_3(2:end); tb_3 = [tip_3(2:end,:), dtRaw_3]; [r3,c1] = find(tip_3==0); if r3 > 1; tb_3 = tb_3(1:r3-1,:); end end if tip_4(1,2) == 4 tip_4(:,2) = []; dtRaw_4 = convSame(diffFilter, tip_4(:)); dtRaw_4 = dtRaw_4(2:end); tb_4 = [tip_4(2:end,:), dtRaw_4]; [r4,c1] = find(tip_4==0); if r4 > 1; tb_4 = tb_4(1:r4-1,:); end end if tip_5(1,2) == 5 tip_5(:,2) = []; dtRaw_5 = convSame(diffFilter, tip_5(:)); dtRaw_5 = dtRaw_5(2:end); tb_5 = [tip_5(2:end,1), dtRaw_5]; [r5,c1] = find(tip_5==0); if r5 > 1; tb_5 = tb_5(1:r5-1,:); end  300  end if tip_6(1,2) == 6 tip_6(:,2) = []; dtRaw_6 = convSame(diffFilter, tip_6(:)); dtRaw_6 = dtRaw_6(2:end); tb_6 = [tip_6(2:end,:), dtRaw_6]; [r6,c1] = find(tip_6==0); if r6 > 1; tb_6 = tb_6(1:r6-1,:); end end if tip_7(1,2) == 7 tip_7(:,2) = []; dtRaw_7 = convSame(diffFilter, tip_7(:)); dtRaw_7 = dtRaw_7(2:end); tb_7 = [tip_7(2:end,:), dtRaw_7]; [r7,c1] = find(tip_7==0); if r7 > 1; tb_7 = tb_7(1:r7-1,:); end end if tip_8(1,2) == 8 tip_8(:,2) = []; dtRaw_8 = convSame(diffFilter, tip_8(:)); dtRaw_8 = dtRaw_8(2:end); tb_8 = [tip_8(2:end,:), dtRaw_8]; [r8,c1] = find(tip_8==0); if r8 > 1; tb_8 = tb_8(1:r8-1,:); end end if tip_9(1,2) == 9 tip_9(:,2) = []; dtRaw_9 = convSame(diffFilter, tip_9(:)); dtRaw_9 = dtRaw_9(2:end); tb_9 = [tip_9(2:end,:), dtRaw_9]; [r9,c1] = find(tip_9==0); if r9 > 1; tb_9 = tb_9(1:r9-1,:); end end if tip_10(1,2) == 10 tip_10(:,2) = []; dtRaw_10 = convSame(diffFilter, tip_10(:)); dtRaw_10 = dtRaw_10(2:end); tb_10 = [tip_10(2:end,:), dtRaw_10]; [r10,c1] = find(tip_10==0); if r10 > 1; tb_10 = tb_10(1:r10-1,:); end end  301  if tip_11(1,2) == 11 tip_11(:,2) = []; dtRaw_11 = convSame(diffFilter, tip_11(:)); dtRaw_11 = dtRaw_11(2:end); tb_11 = [tip_11(2:end,:), dtRaw_11]; [r11,c1] = find(tip_11==0); if r11 > 1; tb_11 = tb_11(1:r11-1,:); end end if tip_12(1,2) == 12 tip_12(:,2) = []; dtRaw_12 = convSame(diffFilter, tip_12(:)); dtRaw_12 = dtRaw_12(2:end); tb_12 = [tip_12(2:end,:), dtRaw_12]; [r12,c1] = find(tip_12==0); if r12 > 1; tb_12 = tb_12(1:r12-1,:); end end if tip_13(1,2) == 13 tip_13(:,2) = []; dtRaw_13 = convSame(diffFilter, tip_13); dtRaw_13 = dtRaw_13(2:end); tb_13 = [tip_13(2:end,:), dtRaw_13]; [r13,c1] = find(tip_13==0); if r13 > 1; tb_13 = tb_13(1:r13-1,:); end end end function out = convSame(f, data) data = [data((max(size(f))/2):-1:1);data;data(end:-1:end... -(max(size(f))/2)+1)]; out = conv(f, data); out = out((max(size(f))):end-(max(size(f)))); end  302  This second script removes artificial tips, caused by the rare earth magnet tripping the reed switch twice in one pass, from the basal drain tipping bucket data sets. %Script name: removeFastTips_Drains %Author: Modified by Nathan Fretz from Matthew Neuner and Steven Momeyer %Project: Diavik Waste-Rock Project %This function removes all tips less than 1 second apart, which are not realtips, but instead are the magnet tripping the reed switch twice in a %single tip. %Use for large, basal drain tipping buckets. %% function Data_mod = removeFastTips_Drains(var) %Creates an array with the Date in column 1 and Time Between tips (in days)in column 2 L = size(var,1); Data_mod(1:L,2) = 0;  %Pre-allocate new array for speed.  i = 1; %Main row counter. j = 1; %Secondary row counter to keep data in new array contiguous. while i < L+1; %if var(i,2) < 1.16203e-5; %1.4 sec in days. if var(i,2) < 1.157407407407e-5; %1 sec in days. i = i+1; else Data_mod(j,:) = var(i,:); i=i+1; j = j+1; end end %Filter for taking derivatives. diffFilter = [1,-1]; %Now recalculate time between tips. Data = Data_mod(:,1); Data = convSame(diffFilter, Data); Data = Data(2:end); Data_mod = [Data_mod(2:end,1), Data]; %Finds first zero from preallocation to trim final array. [r2,c1] = find(Data_mod==0); Data_mod = Data_mod(1:r2-1,:); end function out = convSame(f, data) data = [data((max(size(f))/2):-1:1);... data;data(end:-1:end -(max(size(f))/2)+1)]; out = conv(f, data); out = out((max(size(f))):end-(max(size(f)))); end  303  This third and final script calculates the volume of outflow that passes through the tipping buckets (the Type III North drain tipping bucket in this example). Tipping bucket calibration equations are documented in Appendix B. %Script name: calcRates_Volumes_TIIINDrain %Author: Modified by Nathan Fretz from Matthew Neuner and Steven Momeyer %Project: Diavik Waste-Rock Project %This script creates 2 arrays: %TIIINDrain_Rates has the Date in column 1, Time between tips (in days) in %column 2, mL/s in column 3, L/d in column 4, L in column 5. %TIIINDrain_DailyVolumes has the Date in column 1 and daily L in column 2. %% function [TIIINDrain_Rates,TIIINDrain_DailyVolumes]... = calcRates_Volumes_TIIINDrain(var) %Convert time between tips in days to seconds. secs = var(:,2).* 86400; %Averages time between tips for two adjacent tips to %remove effects from unbalanced tipping bucket and discrepancy in bucket %volume. i = 1; secs(i,2) = secs(i,1); i=i+1; while i < size(secs,1)+1 secs(i,2) = ((secs(i,1)+secs(i-1,1))/2); i=i+1; end %Convert time between tips to flow rate (mL/s). %Contact Matthew Neuner for equation used for 2007. %Contact Steven Momeyer for equation used in 2008 and 2009. %The Type III North Drain (3BNxdrn15) tipping bucket was replaced on June %19, 2010. %Equation for 2010 data up to June 19, 2010: %y = 223.54x^-1.0421 %Equation for 2010 and 2011 data after June 19, 2010: %y = 586.01x^-1.0329 Flow = secs(:,2).^-1.0329.*586.01; TIIINDrain_Rates = [var(:,1),var(:,2),Flow]; %Add 4th column and convert flow from mL/s to L/d. TIIINDrain_Rates(:,4) = Flow(:,1).*86400.*0.001; %Add 5th column consisting of volume of each tip (L). TIIINDrain_Rates(:,5) = TIIINDrain_Rates(:,2).*TIIINDrain_Rates(:,4); %Calculate the volume of water (L) over a specified peroid.  304  %The interval is specified in minutes. %This gives you a nx3 matrix: date, volume, number of rows included. interval = 1440; %1 day in minutes. mins = round((TIIINDrain_Rates(:,1)- datenum ('Jan 01 2000'))*1440); %Converts data to number of minutes since Jan 1, 2000. mins = ceil(mins/interval); %ceil rounds towards + infinity lastmins = mins(1); x = [0,0]; TIIINDrain_DailyVolumes = []; for i= 2:size(TIIINDrain_Rates,1) if mins(i) ~= lastmins TIIINDrain_DailyVolumes = [TIIINDrain_DailyVolumes;... (mins(i-1)*interval)/1440 + datenum('Jan 01 2000'),x]; x = [TIIINDrain_Rates(i,5),1]; %5 refers to the column containing %flow in L. lastmins = mins(i); else x = x + [TIIINDrain_Rates(i,5),1]; %5 refers to the column %containing flow in L. end end TIIINDrain_DailyVolumes = [TIIINDrain_DailyVolumes;... (mins(i-1)*interval)/1440 + datenum('Jan 01 2000'),x]; TIIINDrain_DailyVolumes(:,1) = TIIINDrain_DailyVolumes(:,1)... -interval/(1440*2); %Subtracts half of interval (in minutes). This is done for graphing %purposes, so that a bar graph is plotted correctly. %Create 6th column of cumulative L. TIIINDrain_DailyVolumes(:,4) = cumsum(TIIINDrain_DailyVolumes(:,2)); %Create 7th column of cumulative m^3. TIIINDrain_DailyVolumes(:,5) = TIIINDrain_DailyVolumes(:,4)./1000; end  305  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0073576/manifest

Comment

Related Items