Distinguish calibration, adjustment and verification
Calibration establishes a relationship between an instrument's indications and reference values under stated conditions. Adjustment changes the instrument or its settings. Verification checks whether specified requirements are met. These activities are related, but calling all of them “calibration” hides important information about what was actually done and what confidence the result deserves.
The International Bureau of Weights and Measures publishes the International Vocabulary of Metrology through the Joint Committee for Guides in Metrology. Its terminology is useful when writing an equipment procedure. A stadium team does not need to turn every routine check into a laboratory exercise, but it should use clear language and avoid claiming traceability where no documented chain exists.
For a moisture probe, a manufacturer-supported check may establish that the instrument is functioning without proving that its generic conversion is ideal for the venue's rootzone. A material-specific calibration may be appropriate for some applications, but it requires a sound reference method and representative samples. Do not improvise a correction from a small collection of convenient readings.
Keep an instrument and method register
Assign every device a stable identifier. Record model, serial number, firmware where relevant, measurement units, calibration information, service history and the person responsible for it. Link each result to a method version. If a team changes drop sequence, sampling depth or moisture conversion, that change belongs in the record even when the instrument itself remains the same.
Use manufacturer instructions as the starting point for care and checks. Devices from METER Group, Delta-T Devices, Spectrum Technologies and Stevens Water Monitoring Systems have different operating constraints; a maintenance procedure for one probe is not automatically valid for another. Inspect housings, connectors and cables, and coordinate checks after disruptive maintenance or prolonged storage.
A practical pre-round checklist
- Confirm device identity, battery condition and units.
- Inspect sensing surfaces, cables and connectors.
- Perform the appropriate documented reference check.
- Confirm the sampling route and method version.
- Check clock settings and recent maintenance notes.
- Record faults before collecting more results.
Compare like with like
Consistency reduces avoidable uncertainty. Use the same location definition, depth reference, preparation and sequence when repeating a test. For operator-dependent methods, train staff together and compare their results under controlled conditions. A persistent difference between operators may reveal a procedural issue rather than a changing surface.
Two instruments need not produce identical values to provide useful evidence, but unexplained differences should not be ignored. Compare them under suitable conditions and consider sensing volume, calibration, material contact and measurement principle. Avoid replacing one instrument with another halfway through a trend without documenting an overlap or otherwise explaining the break in comparability.
ISO/IEC 17025 concerns the competence of testing and calibration laboratories. Where an external laboratory is used, check whether its accreditation scope covers the actual service required; accreditation in an unrelated area is not enough. The United Kingdom Accreditation Service maintains information about accredited organisations. A certificate should be read for scope, uncertainty and conditions, not treated as a universal quality badge.
Treat data handling as part of measurement
Record the time zone and distinguish measurement time from upload time. A logger that reconnects after an outage may transmit old observations together. If they are plotted as though they were all collected at reconnection, the chart misrepresents the field. Document changes to logging interval and any clock corrections, particularly when combining manual and automated records.
Missing data should remain visibly missing. Do not draw a continuous story through a gap without explaining any interpolation. Keep raw observations separately from cleaned or summarised data. If a reading is excluded because a probe was disturbed, preserve the reason and the original value. Good quality control is reversible and understandable to someone who was not present.
The Open Geospatial Consortium's SensorThings API is a real example of a framework for describing observations and sensing systems. A small venue need not implement that standard to benefit from the underlying discipline: identify the device, the observed property, the feature being observed and the time. A spreadsheet can preserve these relationships if its structure is deliberate.
Make uncertainty operationally useful
Uncertainty is not a reason to abandon measurement. It is a reason to match the strength of the conclusion to the strength of the evidence. A small apparent change near the normal repeatability of a method may not justify an immediate intervention. A repeated pattern across independent checks can deserve more attention even when no instrument is perfect.
Agree in advance how suspect results will be flagged, who can approve their use and when expert review is required. Review the monitoring programme after renovations, equipment changes or a season of accumulated experience. The objective is not an unbroken stream of numbers. It is a dependable account of what was measured, what changed and how confidently the team can act.
Final reliability question: can the team reconstruct the path from instrument and method to observation, interpretation and work order? If a link is missing, fix the evidence chain before strengthening the conclusion.