Effluent Monitor Alarm During a Tank Transfer

The proposal should not be accepted as framed. The grab sample result of 4 Bq/L is below the 20 Bq/L authorised limit, but a single spot sample taken 30 minutes after the alarm does not establish that the alarm was spurious. The alarm indicates that the monitor saw a count rate three times normal background, and that signal needs an explanation before the line is returned to service. Possible causes include a genuine activity transient in the transferred liquor, a monitor fault, or interference from the transfer itself.

The correct sequence is to keep the isolation valve closed, investigate the cause, and only then decide whether to resume. Resuming first and sampling afterwards inverts the logic: if the discharge was genuine, the material would already be in the outfall before confirmation. The 40 m3 of remaining hold tank capacity means there is no immediate need to discharge; the campaign can be paused without loss of containment. The operator's concern about delay is an operational pressure, not a safety justification, and the decision should be made by the responsible person under the site's effluent control procedure.

The monitor was calibrated 11 months ago against a 12-month interval, so it is still within its calibration period but close to expiry. That means the calibration status does not by itself invalidate the alarm, but it reduces the margin for dismissing the alarm as instrument drift. The decision should not rest on the calibration date alone in either direction.

Checks should include a functional response check with a check source to confirm the detector responds and the alarm setpoint is correct, a review of the monitor's recent trend data to see whether the background count rate had been drifting upward before the alarm, and a check of the transfer line and hold tank for any source of interference. The grab sample should be re-analysed and, if the result is near a decision level, repeated with a larger volume or longer counting time to reduce uncertainty. If the monitor proves faulty, the fault should be corrected and the calibration repeated before the line is returned to service; if the monitor is sound, the source of the activity transient must be identified. In either case the 11-month calibration status should be recorded as part of the investigation.

The event is an effluent monitor alarm with automatic isolation, which is a reportable-type occurrence under most regulatory frameworks even when no limit is exceeded; the site's reporting procedure should be consulted and the notification made within the required time. The authorised limit of 20 Bq/L was not exceeded by the grab sample, but the alarm itself and the resulting abnormal discharge condition should be recorded and assessed.

The discharge accounting should record the volume transferred before isolation, the activity concentration measured, the resulting activity released if any, and the basis for concluding whether the authorised limit was met. The grab sample result of 4 Bq/L should be recorded with its uncertainty and the time of sampling relative to the alarm. The record should also capture the investigation findings, the corrective actions, and the review of the monitor's calibration and setpoint. If the investigation shows the alarm was genuine, the cause should be traced to its source in the transfer operation and the procedure revised; if spurious, the monitor maintenance history should be reviewed. The event should be entered in the site's occurrence register with a closure date and an owner for each action.