Model Documentation

10.1 General

This section describes documentation of groundwater models as submitted to the regulator for review and potential approval. Specifically, this section describes paper and electronic documentation provided to the regulator and hence the pubic as part of an EA or other permitting process. This section does not address model documentation by the proponent and their consultants. It is recognized that documentation procedures are company specific and that storage of paper and electronic files varies considerable from company to company.

10.2 Overview

There are two types of model documentation:

  • Modelling Report - reports that describing the basis for the model, key assumptions, modelling approach, and the study findings, conclusions, and recommendations.
  • Modelling Archives - a combination of modelling journals, documents on pre- and post-processing data analysis, and modelling data and software program files, such that the model could be re-generated for review and/or further refinement.

10.3 Minimum Requirements of Submitted Documentation

The primary requirements for documentation provided by a proponent to the regulator are clarity, transparency, and comprehensiveness.

The documentation provided by the proponent to the regulators should include sufficient detail to enable an independent party to replicate the modelling.

Documents provided by a proponent to the regulator are publically available. It is recognized that varying levels of public interest may be elicited by such documents. Accordingly, these guidelines recommend that reports be structured to provide for increasing levels of detail. For example, the following sequence of increasing documentation detail and specificity may be used:

  • Executive Summary - generally no more than one to four pages.
  • Technical Report - preferably no more than about twenty to thirty pages.
  • Drawings, figures, and exhibits - large maps, computer output plots, or exhibits that provide specificity of site conditions, model layouts, and resultant model simulations.
  • Technical Appendices - each should be topic specific and as long as is necessary to document the topic.
  • Calculation Packages - each should be topic specific and of sufficient length and detail for the reviewer to independently check equations, calculations, and outcomes.
  • Data Files - computer data files that may be used by an independent reviewer or investigator to rerun codes.

10.4 Modelling Report

10.4.1 Interim Reports

Interim reports during the modelling process provide an opportunity to both communicate study status, issues and preliminary results, and to elicit feedback from regulators. The use of interim reports provides an opportunity to obtain agreement within both the proponent and regulator teams on such items as assumptions, data requirements, and analytical methods.

The requirements and schedule of interim reports should be established as part of the model study schedule.

If delivered to the regulator, such interim reports may become part of the official project record and hence publically available. Accordingly, interim reports should be clearly labeled as draft. Interim report should be made final and/or incorporated into the final report.

Topics of interim reports may include:

  • Definition of model objectives
  • Conceptual model development
  • Completion of predictive models and uncertainty analysis.

10.4.2 Final Reports for Conceptual Models

Table 10-1 (modified from MDBC, 2001) summarizes recommended components and content for a final modelling report. These guidelines recognize that project specifics may dictate alternative report formats and contents. The proponent is urged to arrange the report as deemed necessary and appropriate to communicate work done, conclusions, and recommendations.

Table 10 1: Suggested Model Report Structure and Content (modified from MDBC, 2001)

Item Title Detail (as relevant for the objectives and complexity)
1 Study title Select the title carefully to communicate the project goals, outcomes and/or the modelling objectives to the intended audience. Avoid simply using the site name.
2 Executive Summary Summarize model development, management scenarios assessed, and the findings of the study. Briefly explain how the model was developed and refined. Summarize uncertainties, inadequacies, and possible methods of resolving these (e.g. by fieldwork and/or further model development).
3 Introduction State the project objectives, model purpose, and complexity in specific and measurable terms. Introduce the study area and previous work. Describe the resource management issues.
4 Data Review and Hydrogeological Setting Describe the catchment geology, hydrology, and hydrogeology. Describe the data available. Collate the hydrogeological framework, parameters, stresses, and monitoring, and published information (e.g. literature review of papers, reports, etc.). Provide a brief overview of the natural resource development modeled (e.g. mine plan, pumping etc.)
5 Conceptual Model Describe the current conceptual understanding of the aquifer system. Outline uncertainties and limitations. Describe the following aspects of the conceptual flow model (using tabular and graphical formats):
  1. aquifer types, geometry, and measured properties
  2. groundwater flow regime (direction, gradients)
  3. the spatial and temporal variation in natural recharge and discharge
  4. surface water-groundwater interactions (lakes, streams)
  5. groundwater abstractions (various uses)
  6. general influence of resource development
  7. water balance estimates
Describe the following aspects of the conceptual contaminant transport model (if applicable):
  1. contaminants of concern (CoC)
  2. sources & sinks of CoC
  3. applicable transport processes
6 Numerical Model Describe the conversion of the conceptual model into a numerical model. Describe the numerical methods (using tabular and graphical formats):
  1. code selection (including rationale)
  2. model domain & boundary conditions
  3. model layers & discretization of model grid
  4. recharge & evapotranspiration
  5. internal sinks & sources
  6. model parameterization
  7. simulation period, initial conditions & time stepping
  8. model convergence parameters
  9. methods of model calibration
  10. methods of sensitivity & uncertainty analysis
Describe the numerical methods of transport model If applicable:
  1. selection of transport code & solution scheme (incl. rationale)
  2. boundary conditions for transport
  3. simulation period, initial conditions & time stepping
  4. methods of sensitivity & uncertainty analysis
7 Model Calibration Describe and discuss the qualitative and quantitative measures of calibration performance and sensitivity analysis for the model, including:
  1. water balances, including time series of components of the water budget and annual water balances
  2. iteration residual error
  3. lumped residuals and statistics, scattergram plots, etc.
  4. comprehensive comparisons between measured and modelled:
    • groundwater heads (maps, cross-sections, hydrographs, horizontal and vertical head gradients)
    • groundwater-surface water interaction (spring and river flow hydrographs, plots showing gaining and losing reaches of streams, etc.)
  5. the sensitivity/uncertainty analysis approach and outcomes.
Describe and discuss the qualitative and quantitative measures of calibration performance and sensitivity analysis for transport model (if applicable), including:
  1. mass balances, including time series of components of the mass budget and annual mass balances
  2. comprehensive comparisons between measured and modelled:
    • CoC concentrations in groundwater (maps, cross-sections, time trends)
    • CoC concentrations in internal sinks such as pumping wells, seeps or springs (time trends)
    • CoC mass loading to internal sinks (e.g. wells, drains) or surface water (streams, lakes)
  3. the sensitivity/uncertainty analysis approach and outcomes (Note; this aspect is critical since calibration of transport model is often not feasible).
8 Model Predictions Present the model predictions in response to the resource management options simulated, including:
  1. Predicted changes in groundwater flow system (e.g. maps, cross-sections of heads/drawdown, hydrographs of mine inflow etc.)
  2. Predicted contaminant transport (e.g. contour plans of contaminant plumes, breakthrough curves at compliance points, mass loading vs. time, etc.)
Assess the influence of model uncertainty on predicted groundwater flow and/or contaminant transport.
9 Model limitations Describe and discuss the uncertainties in relation to the conceptual model, and model calibration and prediction simulations, and possible methods of resolving them by subsequent data acquisition, field monitoring, further analysis and/or modelling, and resolution by future construction of mitigative measures.
10 Recommendations Summarize the preferred management scenario, and other study findings. Provide conclusions about the potential impact of natural resource development (or alternative scenarios) on the groundwater system and associated surface water (streams, lakes) w/ acknowledgment of model uncertainty. Recommend management plans and future work programs (e.g. additional field work, modelling etc.).
11 References List references of relevant literature. Consider a summary (possibly in the form of an annotated bibliography) of the key reference papers and reports.
12 Appendices Especially for a medium- or high-complexity model, it is recommended that much of the detailed information (e.g. summary sheets of model output for calibration, sensitivity analysis and prediction) be presented in Appendices in graphical and tabular form. This allows for the body of the report to be written in a lucid style for easy communication of the approaches used and issues addressed.

10.4.3 Final Report for Predictive Model

A report on a predictive model should generally follow the guidelines above for a report on a conceptual model. In addition, a report on a predictive model may include the following:

  • Description of the baseline conceptual model that is the basis of the predictive model
  • Details of the mine project and works that give rise to the need for a predictive model
  • Description of the predictive model
  • Discussion of how the predictive model is derived from, based on, or deviates from the conceptual model
  • Plans for monitoring the groundwater to ascertain if and how the new mine works are affecting groundwater
  • A description of mitigative measures that may be implemented if monitoring indicates a significant deviation from model predicted conditions

10.5 Monitoring Plans

If the report on the conceptual baseline model and/or the report on a predictive model recommends additional and/or ongoing monitoring, details of the monitoring should be included in the model report or in a separate, stand-alone monitoring report. At the very least, a monitoring report (plan) and/or sections of a model report dealing with monitoring, should include the following:

  • Reasons for proposed monitoring or data gathering.
  • Description of monitoring facilities
  • Standard procedures for using (measuring) the monitoring facilities
  • Schedules for preparation and submission of reports on the monitoring to regulators
  • Measured trigger levels that may give rise to a need for installation of mitigative measure or the need to undertake remedial work. A trigger level is the value of a measured parameter that may be compared to the value of the parameter as predicted by a model.

10.6 Model Review

Peer reviews should be documented. A typical peer review report may include descriptions and discussion of the following:

  • Model objectives
  • The adequacy of the conceptual model
  • The adequacy of the mathematical model
  • Calibration sufficiency
  • Predictive models reasonableness
  • Sufficiency of the uncertainty analysis

10.7 QA/QC

QA/QC procedures and implementation should be documented. The QA/QC documentation may include descriptions of:

  • Checks on datasets used for model inputs. Data entry or interpretation errors can be identified.
  • Checks on model construction. Construction and application of the mathematical model can be verified. For example, boundary conditions can be checked for entry errors, model grids can be checked for holes or areas of unintended high density, and verification that calibration points are input and used correctly.
  • Checks on model results. Results should be checked for plausibility. Often, during review of model results, model setup errors can be identified that were overlooked during model construction. Checks can include model mass balance, water table contours and hydraulic head measurements. Maps or figures of model results can be used to assess spatial variability in model results or identify possible boundary condition errors.

10.8 Model Archive Documentation

Model archive documentation need not by submitted by the proponent to the regulator. Proponents should recognize however, that in the event of litigation, such documentation is discoverable and should accordingly be prepared to a high standard and kept safe for future reference.

Model datasets and results files should be archived in a way to allow ease of access at later dates. Proper archiving of key modelling input/output files can be important for (i) future follow-up work and/or (ii) potential review during a modelling audit.

It is good practice to use a model journal. Such a journal can provide both a record of what has been completed and a tracking method to locate appropriate files at a future date, if necessary. The journal should be used as a method to track the progression of the modelling process. As such, it is useful to keep concise notes on how the process evolved, inputs received or generated, and model output files.

A good record of model runs, which should be named in a logical manner, provides a useful resource when addressing questions or determining the cause for surprising or incorrect results. This is particularly important during the calibration and uncertainty analysis phases, but also important at other phases. The journal should include file names and locations of datasets, model inputs and model output files.

Review Questions

  1. Which of the following best states the purpose of model documentation?
    1. To identify problems with the conceptual model.
    2. To present graphical outputs of the mathematical model.
    3. To communicate model methods, assumptions, results and limitations in a clear and transparent manner.
    4. To describe predicted effects on the VEC of interest.
    5. All of the above.
  2. Best practices for model documentation include:
    1. Executive summary, introduction, data review and hydrogeological setting, conceptual model, numerical model, model calibration, model predictions, model limitations, conclusions and recommendations.
    2. Executive summary, conceptual model, model predictions, conclusions and recommendations.
    3. Executive summary, introduction, data review and hydrogeological setting, conceptual model, numerical model, model calibration, conclusions and recommendations.
    4. Executive summary, introduction, data review and hydrogeological setting, conceptual model, numerical model, model predictions, conclusions and recommendations.
    5. None of the above.
  3. Documentation on model review & QA/QC should be included because:
    1. It is best practice.
    2. Proper review and QA/QC can avoid setbacks upon completion of a modelling study.
    3. Documentation will ensure that the appropriate model methods have been used.
    4. A and B.
    5. A, B, and C.
  4. Plans for additional or on-going monitoring should focus on data collection only for improving model calibration.
    1. True.
    2. False.
  5. Proper model archive documentation should be maintained because:
    1. It will be reviewed by regulators.
    2. It may be required to allow for future follow-up work.
    3. It may be required for a model audit.
    4. B and C.
    5. A, B, and C.

Proceed to Section 11: Model Review