MEASURE. VALIDATE. INTERPRET.
Instrument Interpretation for Monitoring Data
RAUZ explains what monitoring instruments measure, how common data issues appear, and how inclinometer, piezometer, settlement, load, tilt and survey records should be interpreted in engineering context.
Instrument Interpretation
A reading is evidence. Interpretation explains what the evidence can support.
Instrument interpretation is the structured engineering review of what a monitoring system actually measures, whether the record is credible, how the trend changes with time and location, and whether the pattern is consistent with the documented ground, groundwater, structure, environment and project sequence. The instrument provides an observation; the project context gives that observation meaning.
Know the measured quantity
Displacement, tilt, pore pressure, head, strain, load, crack opening, settlement, temperature and survey coordinates are not interchangeable. Interpretation starts with the parameter, units, geometry, datum, sign convention and measurement method.
Check whether the record is usable
Review baseline, zero reading, timestamps, calibration or verification status, data gaps, step changes, reference stability, maintenance history and any corrections applied before treating an apparent change as physical behaviour.
Connect the pattern to mechanism
Compare magnitude, rate, depth, spatial coherence and timing with construction events, groundwater, weather, neighbouring instruments, survey evidence and the documented engineering mechanism.
Interpretation Framework
Move from raw reading to engineering finding in six visible steps.
The sequence below keeps measurement, data quality and engineering inference separate. It is deliberately vendor-neutral: the same logic can be applied to manual readings, automated loggers, survey systems or third-party monitoring platforms.
Question
What behaviour is the monitoring intended to detect: wall movement, pore-pressure response, consolidation, structural rotation, load transfer, crack movement, slope deformation or another defined mechanism?
Traceability
Which instrument, location, depth, orientation, reference system, logger, processing route and baseline produced the plotted value?
Data quality
Is the series complete and credible? Check abrupt jumps, flat-lines, missing periods, re-zeroing, sensor replacement, reference movement and corrections.
Pattern
Review cumulative change, rate, acceleration, reversals, depth profile, spatial distribution and whether adjacent measurements behave coherently.
Context
Relate the timing to excavation, tunnelling, dewatering, loading, rainfall, temperature, operations, maintenance and other documented events.
Finding
Separate what is observed from what is inferred. Record uncertainty, alternative explanations, missing evidence and the next verification or engineering action.
Instrument Families
Interpret each instrument according to what it can actually observe.
The same visual pattern can mean different things in different measurement systems. The checks below are interpretation prompts, not universal acceptance criteria.
Inclinometers and in-place inclinometers — lateral deformation with depth
What they show: changes in inclination along a borehole or structural profile that are processed into incremental and cumulative displacement relative to a baseline.
Interpret: depth of deformation, shape of the profile, cumulative movement, rate of change, repeatability and whether the apparent movement localises where the expected mechanism could occur.
Check first: casing orientation and depth registration, baseline survey, probe or IPI identity, stable reference zone where applicable, maintenance/replacement history and any processing or sign-convention change.
Useful cross-checks: wall or ground survey, adjacent inclinometers, excavation depth, support sequence, groundwater and observed ground behaviour. GEOKON’s official inclinometer guide emphasises that the raw measurement is tilt and that processing is needed to derive displacement profiles.
Piezometers and standpipes — pore pressure, pressure head and groundwater response
What they show: hydraulic conditions at the instrument location or response zone. The precise interpretation depends on instrument type, installation and reference elevation.
Interpret: change from baseline, rate, drawdown or recovery, depth-dependent response, lag, seasonal behaviour and relationships with dewatering, recharge, rainfall, excavation or loading.
Check first: datum, elevation, units, instrument type, saturation/installation records where relevant, zero/reference condition, venting or barometric considerations where applicable to the instrument system, and whether the plotted quantity is pressure or converted head.
Useful cross-checks: nearby piezometers, standpipes, pumping records, rainfall, water levels, seepage observations, excavation sequence and settlement. FHWA guidance describes piezometers as instruments for pore-water-pressure or head monitoring and uses them alongside settlement and deformation measurements in soft-ground embankment monitoring.
Settlement points, plates and levelling systems — vertical movement and rate
What they show: vertical displacement of a monitored point or layer relative to a defined reference.
Interpret: cumulative settlement or heave, rate, time-dependent behaviour, spatial pattern and the relationship with loading, fill placement, excavation, consolidation, groundwater or construction stage.
Check first: benchmark/reference stability, survey method, reading precision, datum changes, plate/rod integrity, baseline date and whether movement is absolute or relative.
Useful cross-checks: levelling benchmarks, survey prisms, pore pressure, load history, fill height, adjacent settlement points and InSAR where geometry and resolution are suitable.
Tiltmeters, total stations, prisms and GNSS — rotation and spatial movement
What they show: tilt or changes in position, depending on the system. Rotation should not be treated as identical to translational displacement.
Interpret: direction, persistence, spatial coherence, movement of the control network, repeatability and whether the magnitude is meaningful relative to the system’s verified performance.
Check first: reference-point stability, line of sight, prism or target condition, instrument setup, atmospheric/environmental influence, satellite geometry or processing quality where relevant, and maintenance events.
Useful cross-checks: manual survey, levelling, inclinometers, adjacent targets and construction records. Crossrail’s published monitoring review documents how obstructions, damaged or dirty reference prisms, train movements and vibration could create problematic automated survey readings.
Extensometers and crack/joint meters — relative displacement
What they show: relative movement between anchors, fixed points or opposite sides of a crack or joint.
Interpret: opening/closure or extension/compression trend, rate, reversals and whether several measurement points define a coherent deformation zone.
Check first: anchor/reference stability, installation geometry, zero, range, temperature effects where relevant, mechanical damage and whether the monitored feature itself has changed.
Useful cross-checks: visual inspection, survey, structural or geological mapping, neighbouring displacement sensors and construction/loading history.
Load cells and strain gauges — force, strain and structural response
What they show: load or strain according to the sensor, installation and calibration model. Strain readings may require structural assumptions before they are translated into force.
Interpret: load transfer, load distribution, trend, hysteresis, temperature response, seating effects and consistency with the installed support or structural system.
Check first: valid zero, calibration, bearing/contact conditions, installation configuration, temperature, jack calibration where relevant and whether the field configuration matches the calibration assumptions.
Useful cross-checks: applied jacking records, support installation stages, strain at other positions, structural movement and retaining-wall deformation. GEOKON’s official load-cell guidance shows why field configuration, bearing, friction and zero reading can materially affect interpretation.
Environmental and climate sensors — context for physical response
What they show: rainfall, temperature, water level, moisture or other environmental parameters relevant to the monitored system.
Interpret: whether recurring or event-based environmental changes coincide with instrument response. Correlation is useful evidence but is not, by itself, proof of causation.
Check first: sensor siting, timestamp alignment, data continuity, unit consistency, maintenance and whether the environmental measurement is representative of the monitored asset.
Useful cross-checks: pore pressure, settlement, tilt, crack movement, seasonal operating records and independent weather or hydrological observations. FHWA seasonal-monitoring guidance explicitly combines climate and internal material-response measurements to study cause-and-effect behaviour.
Error Patterns & QA/QC
An abnormal graph is a reason to investigate, not an automatic mechanism.
Data quality problems can mimic real movement. The table below lists common patterns and the first questions a reviewer should ask before accepting an engineering explanation.
| Observed pattern | First checks | Interpretation caution |
|---|---|---|
| Single sudden jump | Timestamp, re-zero, probe/system change, logger event, reference movement, manual re-read. | A step change can be physical, but it can also come from a measurement or processing change. |
| Progressive drift | Calibration/verification history, temperature, reference stability, neighbouring sensors, maintenance. | Slow sensor or reference drift can resemble genuine long-term deformation. |
| Flat-line or frozen value | Communication status, power, logger, sensor status, update interval, data pipeline. | No apparent movement is not evidence of stability if the measurement stopped updating. |
| Cyclic pattern | Temperature, tides, operations, rainfall cycles, pumping schedules, daily/seasonal timing. | Repeated cycles should be separated from secular movement before rate is interpreted. |
| One sensor disagrees with neighbours | Instrument health, location, local mechanism, baseline, geometry, independent measurement. | Disagreement may indicate a local response or a data problem; neither should be assumed. |
| Trigger crossing after baseline change | Approved baseline, effective date, offset applied, historical traceability, trigger definition. | A threshold comparison is only valid when measurement and trigger use compatible references. |
| High-frequency spikes | Obstruction, vibration, environmental interference, resection/control quality, filtering method. | Automated systems can produce dense data without every value carrying equal engineering meaning. |
Cross-Source Validation
Confidence increases when independent evidence tells a compatible story.
No rule says every instrument must agree numerically. Different systems measure different quantities, locations and geometries. The purpose of cross-checking is to test whether the observed pattern is physically compatible.
| Engineering question | Primary evidence | Useful corroboration | Key caution |
|---|---|---|---|
| Is a retaining wall moving laterally? | Inclinometer / IPI profile | Survey prisms, adjacent inclinometers, support loads, excavation stage | Compare direction, depth and timing rather than expecting identical values. |
| Is dewatering influencing settlement? | Piezometer / groundwater record | Settlement points, levelling, pumping logs, construction chronology | Coincident trends support a hypothesis but do not prove a mechanism without project context. |
| Is an apparent survey movement real? | ATS / prism / GNSS | Manual survey, levelling, tiltmeters, control-network checks | Reference or line-of-sight problems can move the calculated result without moving the asset. |
| Is a local ground movement part of a wider pattern? | Ground instruments | InSAR, GNSS, survey, neighbouring points | InSAR line-of-sight geometry and spatial resolution differ from point sensors. |
| Is a support load change consistent with excavation response? | Load cell / strain gauge | Wall deflection, support installation records, temperature, excavation depth | Field load interpretation depends on installation, calibration and structural assumptions. |
Project Context
The same reading can carry different meaning in different ground, climate and construction conditions.
A global knowledge page cannot prescribe one geology, groundwater level or climate model. Project interpretation should use verified site information and the actual monitoring objective.
Geology & strata
Soil and rock profile, weak layers, fill, weathering, discontinuities, compressible deposits and stiffness contrasts can control where deformation occurs and which instrument depths matter.
Groundwater & hydraulics
Pore pressure, drawdown, recharge, seepage, tides or seasonal groundwater changes may alter effective stress and can be relevant to settlement, heave and slope behaviour.
Climate & environment
Rainfall, temperature, freeze/thaw, solar heating, humidity or river/reservoir levels can affect the ground, structures, sensors and reference systems. The relevant factor depends on the project.
Construction sequence
Excavation depth, tunnel advance, dewatering, loading, backfilling, grouting, piling, demolition and support installation give the time axis needed to test cause-and-effect hypotheses.
Asset geometry
Instrument location, orientation and distance from the expected mechanism determine what part of the response is observable. A local sensor should not be assumed to represent the entire asset.
Measurement geometry
Survey datum, control network, inclinometer axis, GNSS solution, InSAR line of sight and instrument reference define how physical movement is projected into a reported value.
Official Lessons
Published monitoring practice shows why interpretation must account for accuracy, environment and purpose.
The examples below come from official public sources. They are not RAUZ projects and do not imply endorsement or commercial relationships.
Dense automated data can make instrument error more visible.
Crossrail’s review of three station monitoring programmes reports automated inclinometer error of about ±2 mm over a day on one deep shaft installation. With relatively small movement between excavation stages, significant early effort went into distinguishing instrument behaviour from project movement. The lesson is not that automation is poor; it is that accuracy, expected movement and reading frequency must be interpreted together.
Reference and site conditions can create apparent movement.
The same Crossrail paper describes automated survey spikes associated with damaged or poor reference prisms, dirt, blocked sight lines and railway vibration. Manual checks were valuable during particular works. Interpretation therefore includes the control system and operating environment, not only the target point.
Field configuration matters when load is inferred.
GEOKON’s published load-cell interpretation guidance explains that jack pressure and calculated cell load can differ materially, and discusses field calibration, bearing-plate design, friction, temperature and zero readings. A load value cannot be divorced from how the sensor was installed and calibrated.
Immediate anomaly review and deeper periodic review serve different purposes.
U.S. Bureau of Reclamation Design Standards describe prompt review for noteworthy anomalies and separate periodic in-depth evaluation. The guidance also states that instrumentation data are most useful when the reasons for installing the instruments are understood.
RAUZ Method
Interpret the project’s existing evidence without forcing a new hardware stack.
RAUZ sits above the measurement layer. The review can start from data produced by the client’s existing monitoring contractor, survey team, logger, platform or remote-sensing source, then focus on credibility, trend, cross-source consistency and engineering meaning.
- Vendor-neutral inputs: work from existing data exports, reports and platforms where the required metadata are available.
- QA/QC before interpretation: test continuity, baseline, reference, units, metadata, anomalies and processing history first.
- Cross-source review: compare instruments, survey, groundwater, environmental data, project events and InSAR where technically appropriate.
- Observation vs inference: state what the data show separately from the engineering mechanism proposed to explain it.
- Traceable limits: record missing evidence, uncertainty, alternative explanations and conclusions that require local or project-specific authority.
- Decision-focused output: define what should be verified, monitored, reviewed or escalated next rather than producing another graph for its own sake.
Instrument anomaly review
Investigate a jump, drift, baseline concern, conflicting instrument, trigger exceedance or unexplained change in rate.
Monitoring intelligence
Review agreed datasets on a defined schedule with data-quality observations, trends, anomalies, context and follow-up priorities.
Second technical view
Review whether submitted monitoring conclusions are supported by the available records, while keeping contractual and statutory roles explicit.
Official References
Public technical sources used for this instrument-interpretation guide.
These sources are provided for technical context. External products, organisations and projects remain the property of their respective owners and are not presented as RAUZ projects or endorsements.
Frequently Asked Questions
Instrument interpretation questions.
How do I know whether an abnormal reading is real movement?
Should monitoring data be interpreted against a universal threshold?
Why can two nearby instruments show different movement?
How should rainfall or temperature be used in interpretation?
Can RAUZ interpret data from instruments it did not install?
Can InSAR replace ground instruments?
What should I send RAUZ for an initial instrument-interpretation discussion?
Start With the Evidence
Have an instrument trend that does not yet have a clear engineering explanation?
Send a representative dataset, monitoring report, instrument register or trigger issue. RAUZ can first define whether the question needs a focused data diagnostic, an independent review, recurring monitoring intelligence or a broader cross-source assessment.
Useful first material
Instrument type and location, monitoring objective, baseline, sample data, trigger criteria, current concern, project stage, construction chronology and any relevant ground, groundwater or environmental information.
What the first review can define
Which records are credible, which checks are still required, what comparisons are technically meaningful, what remains uncertain and what follow-up should be discussed with the responsible project team.
Author: RAUZ Technical Team · Technical review: Dr. Xuefeng (Jason) Nong · Published: 3 October 2026 · Founder & technical background