Investigations

This section contains summaries of reports on drinking water
quality incidents and investigations.

Report on drinking water quality incident at:
Gorthleck Water Treatment Works in July 2007

Scottish Water Executive Summary

Gorthleck WTW is a colour removal (nanofiltration) membranes treating river water. Disinfection is achieved by dosing a solution of Sodium Hypochlorite followed by pH correction using limestone chips. The treated water is pumped to a 750 m3 Clear Water Storage Tank (CWT). The plant is designed to produce 17.5 m3 of treated water per hour supplying a population of 471. The final sample tap for the WTW is downstream of the CWT. The site has mains power and telemetry.

On Thursday 19th July the programmed operator site visit did not occur as planned due to the operator being called away for personal reasons. On Saturday 21st July at 1901 hrs the Standby Operator received a low treated water chlorine alarm, the Operator recalls this alarm clearing whilst on the call to the OMC, no action was taken. At 2310 hrs the Standby Operator received a low inlet raw water flow alarm. At 0036 hrs on Sunday 22nd the Standby Operator arrived on site. He checked control room for CWT tank depth and Supply Water Chlorine residual both of which were satisfactory. The Standby Operator then shut the plant down on a controlled shut down to allow him to manually clean the inlet strainer. The plant was then restarted. Crucially the level of sodium hypochlorite in the storage tank was not checked. The Operator left site at 0052 hrs believing that all was working as normal, no alarms were present and control panel was all showing healthy. As the Cl2 tank was already below the alarm point it would not alarm again.

On Sunday 22nd at 1752 hrs a low inlet flow alarm received and passed to the Standby Operator. Standby Operator confirmed that he was not fit to attend site as he had been working at Loch Ashie WTW from 04:33 until 10:49 on that day. A second Operator was authorised to attend site and arrived at 1900 hrs and started to check the WTW starting at the raw water inlet. At approximately 1930hrs it was discovered that the sodium hypochlorite storage tank was empty. The Operator immediately took corrective action and at 2217hrs contacted the Standby Escalation Team Leader who in turn contacted the Public Health Team. The water into supply chlorine trend shows that water to supply remained above the emergency action limit for low chlorine at all times during and after this event and it was decided that no further action was required.

Root cause of the loss of chlorine dosing was due to the Sodium Hypochlorite tank not being filled by the site operator on the Thursday 19th July, resulting in the tank being fully empty by Sunday 22nd July. The length of time of the event was extended due to the standby operator being focussed on low inlet flow and water to supply chlorine and not noticing that the sodium hypochlorite storage tank was very close to empty at the time of his visit at midnight on Saturday 21st.

Actions being taken in response to this incident include:

• One to one interviews have been carried out with the operators involved.
• Alter the setting on the sodium hypochlorite low level alarm and link it to telemetry to ensure that there is always 48 hrs capacity.

DWQR Assessment on the incident

The DWQR assessment of this disinfection failure was that it was caused by the failure to fill up the disinfectant tank when it was almost empty. This task should have been done by the normal operator on 19 July but he was called away for personal reasons and it would appear that no other member of staff was asked to carry out his duties in his absence . A low chlorine alarm on 21 July apparently subsequently cleared itself and so was not investigated. A further low chlorine alarm later the same day on 21 July was not passed to the operator as he had confirmed that he was already due to attend the site to investigate a low raw water flow alarm. Having not been told about the low chlorine alarm and with chlorine residuals being adequate, the operator did not think to check the disinfectant levels. It was therefore not until the early evening on 22 July when the operator was investigating a further low raw water flow alarm, that he noticed the low level of disinfectant. Appropriate remedial action was then taken . It was fortunate that the chlorine residual never fell below the low Emergency Action Level.

DWQR notes the actions being taken by Scottish Water as a result of this incident. These include interviewing the operators, linking an appropriate low level alarm for the disinfectant to telemetry and investigating the feasibility of an auto shut down facility for a low treated water chlorine residual.