Please Note: GEOKON will be closed Thursday, November 28 and Friday, November 29 for the Thanksgiving Holiday. Normal business hours resume on Monday, December 2.

  Loading... Please wait...

Session 04

Geotechnical Instrumentation Data Analysis Tips and Tricks
(Part 2 • Data Presentation, Interpretation and Reporting)

Image of data graph.
Posted November 24, 2021

Every good monitoring program consists of clear presentation of instrumentation data alongside other site metrics to present a cause-and-effect picture of how the site is performing.

The following are some tips to ensure that instrumentation data is presented in the most clear and concise manner.

Data Presentation

  • Plots to assist data screening
    • lines of best fit to smooth out data (remove outliers)
  • Plot data vs time for full data set
    • to assess data quality and show trends
  • Plots showing observed vs predicted behavior
    • plotted regarding time and on same axis
  • Image of plot showing data vs time
  • Plots comparing measurements and observations
    • often form part of data interpretation
    • plotted regarding time and on same axis
    • e.g., fill placement, pond level, rainfall, and pore water pressure
  • Image of plot comparing measurements
  • Plots to examine cause and effect relationships
    • e.g., pore pressure increase and settlement as a function of fill placement
  • Image of plot to examine cause and effect relationships
  • Determine the magnitude of change in the monitoring parameters required to validate performance of the structure
    • and verify that the proposed instruments can detect that magnitude of change
  • Plot error bands on all data sets to assist in interpreting actual change in conditions vs possible sensor errors
  • Image of plot showing IPI cumulative displacement
  • Can the instruments you selected be sampled by the proposed system at the frequency required by your monitoring plan?

Data Interpretation

  • Qualified personnel
  • Follow a pre-determined procedure
  • Do not delay interpretation
    • data collection, processing and interpretation influence each other
  • Create plots with meaningful interpretation in mind
    • e.g., units from a MEMS IPI sensor can be displayed in:
      • Raw data digits
      • Acceleration in g
      • Tilt in radians
      • Tilt in degrees
      • Displacement in mm/m or in/ft
  • First reaction is to reject unanticipated readings and/or blame the sensor, but data may be real (and be important)
  • Consult with manufacturer to better understand performance of the instrument in question
  • Ask “Can I think of a hypothesis that is consistent with the data”
    • check reading correctness to assess data validity
      • zero readings, gage factors, temperature corrections, barometric pressure, datalogger programs/power
    • check raw data
    • check installation records | difficulties encountered during install
    • site and environmental conditions
      • Construction, rainfall, headpond level

Compare Instrumentation data with design models

  • Before reviewing data get a feel for what the values should be based on the design model and existing conditions
  • How sensitive are model results to input parameters (parametric studies should be undertaken to show a range of expected results based on variation in soil type, water level, temperature etc.)
  • Are there any localized conditions that could result in variations from the modelled section?
  • Image showing VW Piezometer installation location and predicted water level

VW Piezometer installation location and predicted water level | Reference: www.geostudio.com

Screen the data for expected site conditions vs. actual site conditions

  • Before reviewing instrumentation data make sure the site conditions and change in those conditions with time is understood
  • Construction data is often not synced to instrumentation data making it difficult to determine the loads the instruments are subjected to at the time of the readings
  • There can be disconnects between the engineer(s) reviewing the instrumentation data and the designer/construction inspector
  • Image showing Upstream constructed dam (McLeod and Bjelkevik 2017)

Reference: www.klohn.com/blog/best-practices-for-tailings-dam-design

Reporting

  • Conclusions, drawn from interpretations must be:
    • Reported and submitted
      • to those responsible for implementation of data on a regular schedule
  • Final report to include
    • introduction, why are we monitoring
    • any design and construction information relative to monitoring program
    • summary of monitoring program plan
    • description of instruments and readouts and calibration procedures
    • plans and sections
    • surface and subsurface geotechnical data
    • procedures for data collection, processing, presentation, and interpretation
    • observed behavior with measured data and influencing factors and predicted behavior
    • conclusions, discussion, and remedial action
    • assessment of monitoring program along with suggested improvements

‹ Session 03Session 05 ›

top