VERIFY. TRACE. TRUST THE DATA.
Monitoring Data Quality & QA/QC Platform
RAUZ validates monitoring data across geotechnical, environmental and agricultural systems through QA/QC, traceable checks, cross-source review and engineer-reviewed data-quality assessment before interpretation.
Platform · Data Quality & QA/QC
Trust the monitoring evidence before interpreting the trend.
Monitoring data QA/QC is the structured process of checking whether observations are complete, correctly identified, internally consistent, physically plausible, traceable and fit for the decision they are expected to support. RAUZ applies these checks before data moves into trend interpretation, alerts, reporting or independent review.
Know where every value came from.
Instrument identity, location, units, timestamps, reference system, baseline and acquisition route remain attached to the observation rather than being lost in a final chart.
Find records that deserve review.
Rules can screen missing records, duplicates, flat-lines, abrupt steps, stale data, unexpected ranges and configuration inconsistencies without silently deleting the raw record.
Check whether the wider evidence agrees.
A questionable trend can be compared with adjacent instruments, survey, groundwater, weather, operational events, construction sequence or remote-sensing evidence where those sources are relevant.
Separate a data flag from a technical conclusion.
The platform can organise and prioritise anomalies; the engineering or environmental meaning remains subject to the applicable project context and human technical review.
Quality Model
One “valid / invalid” flag is not enough for monitoring data.
A monitoring record can pass one test and fail another. RAUZ therefore treats data quality as a set of separate dimensions that can be reviewed independently and then interpreted together for the intended use.
Completeness
Were the expected records received for the required period, locations, instruments and parameters?
Consistency
Are identifiers, units, sign conventions, timestamps, baselines and transformations applied consistently across the dataset?
Plausibility
Is the value credible for the measurement method, instrument range, site conditions and observed physical system?
Continuity
Do gaps, resets, maintenance events, re-zeroing or reference changes explain apparent discontinuities in the time series?
Traceability
Can a reviewer reconstruct the raw observation, processing steps, exclusions, corrections and later decisions?
Fitness for use
Is the available quality sufficient for the actual decision, alert, report or engineering interpretation being made?
QA/QC Checks
Screen the data chain, not just the final number.
The useful checks depend on the source and project. The RAUZ platform can structure a repeatable review around the most common failure modes while retaining the raw evidence for later inspection.
Missing, late or duplicated data
Identify unexpected gaps, duplicate timestamps, stale observations, out-of-sequence records and periods where communication loss may be confused with physical stability.
Range and plausibility checks
Flag impossible, out-of-range or contextually unusual values using instrument limits and project-specific rules rather than a single generic threshold.
Flat-lines, steps and drift
Detect prolonged constant values, abrupt offsets, resets, progressive drift and other shapes that may indicate a measurement or configuration issue.
Units, sign and scaling
Check whether a channel is mapped to the correct unit, sign convention, gauge factor, coordinate direction, conversion equation and processing route.
Baseline and datum control
Keep baseline changes, re-zeroing, survey-control changes, reference movement and effective dates visible so that apparent displacement is not detached from its reference.
Identity and change history
Link readings to instrument ID, location, depth or elevation, orientation, channel, maintenance, replacement, calibration or verification status and known configuration events.
Neighbouring and independent evidence
Compare related instruments or independent observations where technically appropriate to see whether an apparent change has spatial, temporal or physical support.
Raw vs derived values
Keep calculated rates, filtered series, cumulative displacement, corrections and other transformations traceable to the underlying source values.
Flag, hold, accept or escalate
Use visible review states so that an unresolved anomaly does not silently enter a report or alarm workflow as though it were already verified.
Platform Workflow
From incoming observation to review-ready evidence.
RAUZ keeps automated checks and engineering interpretation as separate stages. The platform can accelerate repetitive QA/QC without allowing automation to hide uncertainty or rewrite the measurement history.
Receive files, exports, APIs or other agreed data routes.
Align identifiers, time, units, references and metadata.
Run repeatable completeness, range and continuity checks.
Assign visible data-quality and review states.
Compare related sensors, events and independent evidence.
Engineer or specialist examines material flags and context.
Use qualified data in analytics, alerts and reporting.
Evidence Review
A flagged reading is a question. Cross-source evidence helps answer it.
RAUZ uses QA/QC to decide what needs attention, then brings the relevant evidence into view. The objective is not to force different datasets to agree; it is to understand whether they are comparable, whether disagreement is explainable, and what remains uncertain.
Repeatability and neighbouring records
Compare earlier readings, adjacent depths or channels, repeated measurements and instrument status before assuming a sudden change is physical.
Survey, GNSS, InSAR or other observations
Where spatial and temporal references are compatible, independent methods can provide useful confirmation or reveal that two datasets are measuring different parts of the system.
Construction, groundwater, weather and operations
Events may explain why a credible change occurred. They should be retained as context without being treated as proof of causation on their own.
Can the platform automatically decide whether a movement is real?
What if two valid datasets disagree?
Does QA/QC mean deleting outliers?
Monitoring Domains
The quality logic is common. The failure modes are domain-specific.
RAUZ applies the same evidence-first platform architecture across ground and infrastructure, environmental monitoring and agricultural systems. The checks are adapted to the physical parameter, instrument, spatial scale and intended use.
Geotechnical and structural monitoring
Typical QA/QC concerns include depth and orientation, survey control, datum stability, baseline continuity, gauge factors, groundwater context, construction events and coherence between related instruments.
Environmental monitoring data
Quality review may need to address calibration or verification, representativeness, sampling location, contamination or interference, completeness, timing and whether the data is suitable for the intended environmental decision.
Growing-environment observations
Sensor placement, soil heterogeneity, irrigation events, microclimate, maintenance, seasonal change and spatial representativeness can materially affect whether a record describes the actual growing environment.
Outputs
Make data quality visible to the people using the data.
QA/QC is useful only when its result survives into the analysis and reporting workflow. Depending on the agreed scope, RAUZ can expose data-quality status alongside the monitoring evidence rather than hiding it in a back-office process.
Use Cases
Where monitoring data quality becomes a project problem.
The platform is most useful when a project has more data than confidence: multiple systems, automated streams, historical files, unexplained jumps or reporting pressure that makes manual checking inconsistent.
An apparent trigger exceedance
A reading crosses a project threshold. QA/QC checks instrument identity, units, baseline, timestamp, continuity, reference behaviour and related evidence before the event is carried into engineering interpretation.
A silent baseline or reference change
A series appears to “reset” after maintenance, survey control change or re-zeroing. The platform can retain the change event and separate historical comparability from the new reference condition.
Automated monitoring with intermittent communications
Missing packets, delayed uploads or stale values can create false impressions of continuity. Data-age and completeness checks help distinguish communications behaviour from physical behaviour.
Different systems report different answers
Survey, logger, contractor spreadsheets and remote sensing may use different time bases, references or processing. The review begins by establishing comparability rather than forcing the datasets into agreement.
Legacy data needs to be reused
Historical Excel, CSV or reports can be brought into a structured QA/QC review before they are used for trend analysis, model calibration, baseline comparison or independent review.
Recurring reports consume engineering time
Repeatable quality checks can be automated so reviewers spend less time finding missing readings and more time on the anomalies, trends and limitations that actually need technical judgement.
Official Technical Context
Data quality is judged against its intended use, not by appearance alone.
RAUZ uses official quality-management and environmental-data guidance as technical context. These references do not mean that every RAUZ project is certified to, governed by or contractually required to follow each document; applicable requirements remain project- and jurisdiction-specific.
QA, QC and Data Quality Assessment
EPA distinguishes broader quality assurance planning from specific quality-control measures and describes Data Quality Assessment as evaluating whether environmental data are of the type, quantity and quality needed for their intended use.
Field methods, data management and auditing
A 2026 USGS quality-assurance plan for water-quality and water-level monitoring covers field methods, data collection, data management, auditing and equipment management to support reliable, useful and defensible data products.
Confidence in measurement results
ISO 10012 specifies requirements for measurement management systems intended to provide confidence in the validity and reliability of measurement results and to make measurement processes fit for purpose.
Sensor-data quality measurement
ISO 8000-220 specifies quality measures for sensor data, including principles for measuring data-quality characteristics and addressing data anomalies.
RAUZ Role
A vendor-neutral quality layer above the measurement system.
RAUZ does not require one sensor brand or one field architecture. Data Quality & QA/QC sits between data acquisition and interpretation, helping clients preserve their existing instrumentation while adding a consistent, traceable review layer across mixed data sources.
Sensor-neutral
Work from suitable outputs from existing monitoring, survey, environmental and remote-sensing systems rather than forcing a replacement hardware stack.
Cross-domain
Apply a common data-quality architecture across infrastructure, environmental and agricultural monitoring while adapting the checks to each physical system.
Traceable
Keep source observations, processing, flags, corrections, limitations and reviewer decisions visible enough for later technical review.
Engineer-reviewed
Use automation to reduce repetitive checking while keeping engineering and environmental conclusions accountable to appropriate human review.
FAQs
Monitoring data quality and QA/QC questions.
The answers below describe the RAUZ platform approach. Project-specific acceptance limits, statutory requirements and sign-off responsibilities must still be defined for the actual appointment.
What is the difference between QA and QC in monitoring?
Can RAUZ run QA/QC on data from an existing monitoring platform?
Does the platform automatically delete bad data?
Can QA/QC prevent false trigger alarms?
Can RAUZ compare monitoring data with InSAR or survey observations?
Does one QA/QC rule set work for every project?
Is RAUZ Data Quality & QA/QC limited to geotechnical monitoring?
Discuss Your Data
Have monitoring data you do not fully trust?
A useful first review can begin with an existing dataset, monitoring report, instrument register or QA/QC procedure. RAUZ can help identify whether the next step is data-quality screening, anomaly diagnostics, cross-source validation, independent review or a recurring platform workflow.
- Sample CSV, Excel or platform export
- Instrument schedule and metadata
- Baseline and reference information
- Trigger or alert criteria
- Relevant site, environmental or construction events
- Current monitoring or QA/QC report