VERIFY. TRACE. TRUST THE DATA.
Geotechnical Monitoring Quality & Data Assurance
GeoSmar applies structured QA/QC to monitoring data, checking calibration, baselines, timestamps, completeness, anomalies and traceability before engineering interpretation, alerts and reporting.
Quality & Data Assurance
Before a monitoring trend is interpreted, the data has to be trusted.
Geotechnical monitoring quality is not a single calibration certificate or a final spreadsheet check. It is a chain of controls that begins with the engineering question, continues through instrument identity, installation, baseline, acquisition and validation, and ends with a traceable record of what was accepted, questioned, corrected or escalated.
Measurement integrity
Is the instrument suitable, identifiable, installed correctly, calibrated or checked as required, and operating within an understood measurement range?
Data integrity
Are timestamps, units, metadata, baseline values, completeness and processing steps controlled so that the record can be reconstructed later?
Engineering validity
Does the apparent movement agree with neighbouring instruments, construction sequence, groundwater response, survey evidence and the expected mechanism?
Decision traceability
Can a reviewer see why a reading was accepted, rejected, remeasured, escalated or used to support a trigger response?
Assurance Framework
A six-stage chain from measurement to decision.
The controls should be defined before data begins to flow. The exact procedure depends on the project, but the sequence below provides a practical structure for monitoring data assurance.
Official context: the U.S. Bureau of Reclamation states that effective instrumentation monitoring depends not only on what is measured, but also on the timely transfer, review and evaluation of data by decision-makers. Bureau of Reclamation — Design Standards No. 13, Chapter 11
Instrument Control
Data quality starts before the first reading.
A monitoring database cannot repair an unsuitable instrument, a weak reference point or an undocumented installation. The assurance plan should define what must be checked for each measurement system and who owns that check.
| Monitoring system | Typical assurance checks | Common review question |
|---|---|---|
| Total station + prisms | Instrument status, control network stability, prism identity, reference points, atmospheric/geometric corrections, resection quality. | Is the apparent movement at the target, or in the control network? |
| GNSS | Antenna and receiver identity, mounting stability, reference solution, satellite geometry, data gaps and processing method. | Is the displacement significant relative to solution quality and site conditions? |
| Inclinometer | Casing orientation, depth reference, probe/system identity, baseline survey, repeatability, checksum behaviour and depth consistency. | Is the profile change coherent with depth and the expected deformation mechanism? |
| Piezometer / VW sensors | Sensor identity, zero/baseline, cable/logger channel mapping, temperature effects where relevant, groundwater datum and response checks. | Is the change hydraulic, construction-related, seasonal, or potentially instrumental? |
| Tilt / crack / extensometer | Orientation, mounting condition, zero reference, temperature sensitivity, physical inspection and logger channel integrity. | Does the local reading agree with independent evidence from the asset? |
| InSAR-derived motion | Product/source identity, reference convention, temporal coverage, line-of-sight interpretation, spatial coherence and comparison with ground data. | Is the satellite-derived trend consistent with the engineering geometry and ground observations? |
Control also includes instrument identity, installation records, checks on operating condition, configuration changes, reference stability and the treatment of data collected when equipment is later found to be outside its expected condition.
Official context: FHWA quality-control guidance documents instrument calibration, calibration checks, reasonableness checks and the treatment of data collected with equipment later found out of calibration. FHWA — LTPP Data Collection Operations
Baseline & Configuration
A baseline is a documented reference condition, not simply “zero”.
Baseline planning should allow the project team to understand repeatability, seasonal or thermal influence, existing movement, groundwater variation and the effect of any construction already underway. A change in baseline can materially change the apparent displacement, so baseline revisions require explicit control.
Before construction
Where programme allows, collect enough pre-works data to establish normal variation and identify unstable instruments or references before construction effects are expected.
At configuration change
Record logger replacement, sensor replacement, re-zeroing, survey-control change, software transformation or any change that can create a step in the series.
After disruption
Power loss, accidental impact, instrument access, excavation around a sensor, flooding or physical repair should prompt a defined verification before continuity is assumed.
- Baseline date and time
- Raw reading retained
- Reference datum defined
- Units and sign convention recorded
- Instrument ID and location linked
- Reason for re-baselining documented
- Reviewer / approver identified
- Effect on triggers assessed
Official project lesson: Crossrail’s Hyde Park tunnelling instrumentation programme reported more than 12 months of baseline monitoring, used to confirm instrument/measurement accuracy and assess thermal and seasonal effects. Crossrail Learning Legacy — Field Instrumentation Lessons
Data QA/QC
Automated checks should find problems early, not hide them.
A useful QA/QC workflow separates raw observations from derived or filtered outputs. Flags should describe what happened to a record rather than silently removing information that may matter later.
Missing or delayed data
Expected-versus-received counts, communication gaps, sensor downtime, logger backlog and periods with no valid observation.
Range & reasonableness
Physical range, engineering range, rate-of-change checks, impossible values, duplicated records and unit/sign inconsistencies.
Cross-sensor comparison
Compare neighbouring instruments, independent survey, groundwater, construction activities and other datasets that should respond together.
Processing history
Retain raw records and identify filters, corrections, offsets, remeasurements, manual edits and reviewer comments.
Outlier detected
Sudden step change
Gradual drift
Trigger exceeded by one instrument only
Official industry context: Leica GeoMoS describes outlier detection, data validation, filtering and automatic remeasurement as part of automated monitoring data acquisition; its help documentation also distinguishes raw and smoothed series. Leica Geosystems — GeoMoS
Trigger Assurance
A trigger is only useful if the data and the action path are both controlled.
Quality assurance for trigger systems extends beyond comparing a reading with a number. The project should define the measured quantity, baseline, processing rule, verification route, notification responsibility, decision authority and required action.
Before the threshold
Confirm units, baseline, trigger definition, data frequency, valid-data rule, escalation contacts and whether the trigger applies to raw, corrected or derived values.
At the threshold
Record the observation, time, instrument status, verification result, related measurements, site condition and who was notified.
After the response
Keep the technical reasoning, instruction, inspection outcome, increased frequency and any approved change to trigger or monitoring strategy.
Official project lesson: Crossrail Stepney Green used project-specific Green, Amber and Red trigger zones linked to calculated lining displacement and defined review/inspection actions. Crossrail Learning Legacy — Stepney Green SCL Caverns
Data Traceability
Every important value should have a recoverable history.
When several contractors, instruments and platforms are involved, data assurance depends on a stable identity and metadata model. A chart without instrument, location, timestamp, units, baseline and processing history is difficult to audit and easy to misinterpret.
Minimum measurement identity
- Project / asset / zone
- Instrument and sensor identifier
- Location and reference system
- Observed parameter and units
- Timestamp and time zone convention
- Raw observation or derived result status
Minimum change history
- Calibration or verification status
- Baseline or zero change
- Filter / correction / offset applied
- Sensor or logger replacement
- Manual edit or invalid-data flag
- Reviewer, date and reason for change
Projects often combine survey, geotechnical, environmental and satellite-derived information. A consistent data model helps prevent the engineering meaning of a measurement from being lost when it moves between systems.
Official standards context: OGC SensorThings API provides a standard way to manage observations and metadata from heterogeneous sensor systems through entities such as Things, Locations, Sensors, Datastreams, ObservedProperties and Observations. Open Geospatial Consortium — SensorThings API
Project & Contract Interfaces
Many monitoring-data failures are interface failures.
A technically capable instrument can still produce a poor project outcome when responsibilities are unclear. Quality and data assurance should be translated into scopes, deliverables and acceptance rules that are visible to the owner, designer, monitoring contractor and reviewer.
| Interface | What should be defined | Why it matters |
|---|---|---|
| Data ownership | Who owns raw data, processed data, metadata, configurations and close-out archives. | Avoids loss of the engineering record at handover or contractor change. |
| Acceptance criteria | Required completeness, frequency, calibration/check status, valid-data rules and review deadlines. | Makes “good data” measurable rather than subjective. |
| Alarm responsibility | Who receives, verifies, escalates and closes an alert, including out-of-hours arrangements. | Prevents a technically correct alarm from failing operationally. |
| Change control | Approval for baseline changes, trigger changes, sensor replacement, filtering and configuration edits. | Protects continuity and prevents unexplained steps in the series. |
| System availability | Communication outage, local buffering, data recovery, maintenance windows and failure notification. | Distinguishes “no movement” from “no data”. |
| Independent review | Access to raw data, QA flags, reports, calculation basis and records needed to reproduce conclusions. | Allows engineering findings to be challenged and verified. |
Project-Specific Context
Data assurance cannot be separated from geology and construction.
This is a global company capability page, so it would be misleading to invent a single “local geology” or groundwater profile. Project-level QA/QC must instead be configured against the actual ground model, construction sequence and asset sensitivity.
Expected mechanism
Stratigraphy, stiffness contrast, weak layers, rock structure, fill, consolidation and known movement mechanisms determine what a credible response may look like.
Pressure & groundwater
Dewatering, recharge, rainfall, tidal or seasonal effects can change pore pressure and displacement. QA review should distinguish plausible hydraulic response from sensor behaviour.
Sequence & timing
Excavation stages, support installation, tunnelling advance, loading, grouting and temporary works provide the time context needed to interpret a trend.
Information GeoSmar would request
Ground investigation and geological model, instrument schedule, installation records, baseline, raw/processed data, trigger plan, construction programme, survey control information, incident history and reporting requirements.
What changes after review
The result may be a clearer QA plan, revised metadata structure, additional verification, a focused diagnostic review, improved trigger workflow, monitoring-plan changes or a defined recurring data-review service.
Official Engineering Lessons
What established monitoring programmes show about data quality.
The examples below are official public sources used as engineering context. They are not GeoSmar projects and do not imply endorsement, partnership or participation by GeoSmar.
Baseline before interpretation
Crossrail’s Hyde Park instrumentation work reported more than 12 months of baseline monitoring, allowing instrument accuracy and thermal/seasonal effects to be assessed before tunnelling response was interpreted.
Threshold linked to action
Green, Amber and Red trigger zones were linked to calculated lining displacement and explicit actions, illustrating why trigger assurance includes both the measurement basis and the response procedure.
Timeliness is part of monitoring effectiveness
Reclamation guidance stresses that data must reach evaluators and decision-makers in time to be reviewed and acted upon; frequency of reading alone is not enough.
Calibration, independent review and QC flags
FHWA’s LTPP data-quality programme documents equipment calibration, independent review of data procedures, automated data checks, missing-data handling and quality indicators for records.
Monitoring Technology Context
Modern monitoring platforms automate checks — engineering governance still matters.
Commercial monitoring platforms increasingly support multi-sensor data collection, outlier detection, validation, filtering, network adjustment, alarms and reporting. These functions are valuable, but the project still needs rules for what is valid, what a trigger means and who can change the data-processing configuration.
Leica GeoMoS
Leica’s official documentation describes third-party sensor connectivity, data validation, outlier detection, filtering, automatic remeasurement and centralised data storage.
Trimble 4D Control
Trimble describes a platform that manages geodetic, environmental and geotechnical sensor data, analysis and alarms to support timely monitoring decisions.
GeoSmar Role
Quality assurance should make monitoring conclusions easier to defend.
GeoSmar focuses on the engineering layer between measurement and decision. The scope can be a one-off review, a data-recovery exercise, monitoring-plan assurance or an ongoing quality-and-intelligence workflow.
Independent QA review
Review monitoring plans, instrument schedules, baseline rules, raw and processed data, trigger logic, contractor reports and anomalies from an independent engineering perspective.
Data diagnostics
Investigate discontinuities, drift, conflicting sensors, suspicious trigger exceedances, missing data, inconsistent baselines and unexplained changes in trend.
Assurance workflow design
Define practical data checks, metadata requirements, review status, escalation workflow, reporting structure and handover requirements around the client’s existing monitoring systems.
FAQs
Quality & data assurance questions.
Is Quality & Data Assurance the same as instrument calibration?
Can GeoSmar review data from systems it did not install?
Should outliers be deleted?
How should a trigger exceedance be checked?
Can one QA/QC procedure be used for every project?
Does GeoSmar certify that monitoring data is correct?
Start a Technical Review
Have monitoring data you do not fully trust?
Send a sample dataset, monitoring report, instrument schedule or QA/QC procedure. GeoSmar can help define whether the next step should be an independent data review, focused diagnostics, a revised assurance workflow or recurring monitoring intelligence.
Official References
Public sources used for this technical discussion.
The external material below is used as engineering context only. Product and project references remain the property of their respective organisations and do not imply endorsement or a commercial relationship with GeoSmar.
Current GeoSmar positioning around monitoring intelligence, QA/QC, trend analysis, independent review and engineering interpretation.
Official GeoSmar pageDesign Standards No. 13, Chapter 11: Instrumentation and Monitoring.
Official sourceCalibration, data-collection controls, independent review, automated checks and data quality systems.
Official sourceLong baseline monitoring and assessment of thermal and seasonal effects.
Official sourceProject-specific trigger zones and defined actions for SCL monitoring.
Official sourceOutlier detection, validation, filtering, remeasurement and multi-sensor monitoring data handling.
Official sourceMulti-sensor monitoring data management, analysis, alerts and stakeholder updates.
Official sourceSensorThings API for observations and metadata from heterogeneous sensor systems.
Official source