Reading path: Coastal field methods →
Record the attempt, not an invented value
When a reading cannot be obtained, enter the attempt in the same sequence as successful measurements. Include the time, location, target, method, and instrument identifier. Mark the value as unavailable or failed using a consistent notation, rather than leaving a blank that could later be mistaken for an omission.
For example: “10:36, pool B, depth at marked point; gauge DG-1 lowered twice, display remained at zero; result unavailable.” This tells a later reader that the field was visited and a method was attempted, even though no valid number was produced.
Describe instrument behavior
Write what the device did in observable terms: flickering display, unstable reading, no response, damaged cable, or a value that failed to settle. Avoid assigning a cause unless it was established. “Display froze after immersion” is an observation; “water damaged the circuit” is an explanation that may not be known.
If you tried a check or alternative, record its sequence and result. A second instrument that gives a stable reading can help distinguish a device problem from a difficult measurement condition, but label which instrument produced each value. Keep the failed attempt visible rather than replacing it with the alternate value.
Explain how the result was handled
If a value appeared but was rejected, preserve it and state the reason. For instance: “12.8 °C shown briefly, rejected because probe was not fully immersed.” If the method was interrupted by a wave or obstruction, say so. This lets later reviewers evaluate whether the rejection was appropriate and prevents a missing value from hiding the field conditions.
Use a marked status such as “not obtained,” “invalid,” or “excluded,” and define those terms in a project note if more than one observer is involved. Do not use one symbol for both a failed attempt and a measurement that was never planned.
Carry the gap into summaries
When building a table, retain the status and reason in separate fields where possible. A failed reading should not be converted to zero, since zero is a measurement value with a different meaning. If the dataset requires a blank or code for unavailable data, preserve the explanatory note and link it to the notebook entry.
A well-documented failure prevents false precision. It also helps improve a future visit by showing which instrument behavior or field condition needs attention.
Have a correction or a source to suggest? Send a note →