EVALUATION OF METHODS AND MODELS USED TO PREDICT WATER QUALITY AT HARDROCK MINE SITES: SOURCES OF UN

1
EVALUATION OF METHODS AND MODELS USED TO PREDICT
WATER QUALITY AT HARDROCK MINE SITES SOURCES OF
UNCERTAINTY AND RECOMMENDATIONS FOR IMPROVEMENT

• Ann Maest, James Kuipers, Connie Travers, and
  David Atkins
• Buka Environmental Kuipers and Associates
  Stratus Consulting, Inc.
• WMAN Conference, Worley, ID
• October 1, 2005

2
Why Characterize and Predict?

• Regulators use characterization and modeling
  information to determine if a mine will be
  protective of water resources during and after
  mining
• Will mine generate acid and contaminants?
• Future environmental liability set bonds
• Cost of remediating mine sites on the National
  Priorities List (NPL) 20 billion
• Recent increases in the prices of precious and
  base metals have triggered increase in new mines
  around the world
• 170 large hardrock mines in US in various stages
  of permitting, operation, closure

3
This Study

• Lays out framework for evaluating methods and
  models used to predict water quality at hardrock
  mine sites
• Makes recommendations for improvement
• Intended audience regulators, citizens, mine
  operators and managers

4
Nature of Predictions

• Forward modeling
• Timeframe of impacts
• Uncertainties
• Regulatory authorities require predictions

5
Study Approach

• Synthesize existing reviews
• Develop toolboxes
• Evaluate methods and models
• Recommendations for improvement
• Outside peer review (Logsdon, Nordstrom, Lapakko)
• Case studies NEPA/EIS Study

6
Characterization Methods

• Method description
• Method reference
• Use in water quality predictions
• Advantages
• Limitations
• Characterization during different phases of mining

7
Sources of Uncertainty - General

• Extent/representativeness of environmental
  sampling
• need more environmental sampling let
  geologic/mineralogic variability dictate extent
  of sampling define geochemical test units

8
Recommended Minimum Samples
9
Sources of Uncertainty Static

• Effect of mineralogy on NP and APP
• Rely on mineralogy more than on operationally
  defined lab tests
• Interpretation of static testing results
• only use as initial screening technique to
  estimate total amount of AGP/ANP

10
Sources of Uncertainty Leach Tests

• Waterrock ratio
• never known definitively 201 too dilute
• Use of unweathered materials
• must start with weathered materials
• Interpretation of results
• may have limited use as scoping tool if use
  weathered rock and evaluate applicability of
  results

11
Sources of Uncertainty - Kinetic

• Particle size
• minimize amount of size reduction for samples
  field/lab discrepancies
• Length of tests
• 20 weeks is too short for kinetic tests, unless
  shown to be AG before then. NPAPP.
• Interpretation of results
• analyze effluent for all COCs use for short- and
  long-term AGP/leaching potential

12
Length of Kinetic Tests
Source Nicholson and Rinker, 2000 (ICARD).
13
(No Transcript)
14
(No Transcript)
15
Modeling Toolbox

• Category/subcategory of code
• Hydrogeologic, geochemical, unit-specific
• Available codes
• Special characteristics of codes
• Inputs required
• Modeled processes/outputs
• Step-by-step procedures for modeling water
  quality at mine facilities

16
Modeling Opportunities
17
(No Transcript)
18
Sources
19
Pathways
20
Processes
21
Sources of Uncertainty - Modeling

• Conceptual model
• Conceptual models are not unique and can change
  over time
• Revisit conceptual models and modify mining plans
  and predictive models based on new site-specific
  information
• Use of proprietary codes
• need testable, transparent models difficult to
  evaluate, should be avoided. Need efforts to
  expand publicly available pit lake models
  (chemistry).
• Modeling inputs
• large variability in hydrologic parameters
  seasonal variability in flow and chemistry
  sensitivity analyses (ranges) rather than
  averages/medians
• Estimation of uncertainty
• Acknowledge and evaluate effect on model outputs
  test multiple conceptual models
• there is considerable uncertainty associated
  with long-term predictions of potential impacts
  to groundwater quality from infiltration through
  waste rock...for these reasons, predictions
  should be viewed as indicators of long-term
  trends rather than absolute values.

22
(No Transcript)
23
(No Transcript)
24
Summary

• Characterization methods need major
  re-evaluation, especially static and short-term
  leach tests
• Increased use of mineralogy in characterization
  make less expensive, easier to use/interpret
• Modeling uncertainty needs to be stated and
  defined
• Limits to reliability of modeling use ranges
  rather than absolute values
• Increased efforts on long-term studies and
  collection of site-specific data over modeling

25
Conclusion

• Predictive modeling is an evolving science with
  inherent uncertainties
• Using the approaches described in this report,
  predictive water quality modeling and site
  characterization information can be reliably used
  to design protective mitigation measures and to
  estimate the costs of future remediation of
  hardrock mine sites.