Investigations

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

Report on drinking water quality incident at:
Marchbank Water Treatment Works in July 2006

Scottish Water Executive Summary

Marchbank Water Treatment Works (WTW) treats raw water abstracted from Megget Reservoir, to supply parts of Edinburgh and West Lothian. The events that transpired between the dates of 25/07/06 and 28/07/06 were a direct result of water being introduced to the potable water system from Rapid Gravity Filter 4A that had been offline since 27/06/06. The filter had been offline, waiting on spare parts to rectify an intermittent fault with the actuated outlet valve.

On 25/07/06, a planned maintenance task to rectify the fault on the valve actuator was undertaken by Scottish Water Electrical & Mechanical (E&M) staff. During the course of this operation, the outlet valve was opened at 11.36 am, allowing the volume of water contained within the filter to enter the treatment process. The volume of water within the filter was approximately 30m3 and took about 6 minutes to enter the process. The Treatment Operator on site immediately instigated an auto-wash cycle of the filter following the above sequence of events. The Operator did not escalate the above event at this point as he considered that the actions taken in terms of instigation of the backwashing cycle and the volume of water involved, did not pose a threat into the final quality of the water leaving the WTW.

The reaction to the initial taste and odour complaints in the evening of 25/07/06 was to investigate the performance and operation of the chloramination system at the WTW. This was based on historical issues resulting to its configuration. Adjustments to the ammonia and chlorine ratio have been ongoing to try and resolve a Nitrite issue within the distribution network. There was also a perception that alterations had been made to the chloramination system earlier that day when in fact there hadn't. Once the root cause was identified by the out of hours staff, a systematic distribution network flushing and scouring programme was immediately agreed and instigated to minimise the impact of the poor quality water. The flushing activity was based on the estimated travel time of water from the WTW.

Based on reducing customer contacts, this exercise was viewed as a success by 08.00am on the morning of 26/07/06. However, following an increase in customer complaints by mid-day of 26/07/06, a full Scottish Water incident team was formed. Subsequent flushing, sampling, media releases and external communications were undertaken under the auspices of the incident structure. Discussions with the Consultant in Public Health Medicine (CPHM) resolved that the issue was one of an aesthetic nature. The incident was finally stood down on 28/07/06.

DWQR Assessment on the incident

Following investigations and review of the available evidence, DWQR concludes that Scottish Water's explanation for the incident, namely the release of "stale" water into supply during the planned work on the outlet valve to Rapid Gravity Filter (RGF) 4A is correct as the most likely cause of the incident. Trace organics sampling and consultation with health professionals indicated that no substances harmful to health were present and that the problem was an aesthetic one.

DWQR investigations show that Scottish Water does not have a detailed procedure specifically for working on water treatment assets whilst they are out of service. Scottish Water's own report into this incident highlights this omission. Such a procedure should have contained measures to minimise the impact of the work on water quality. The additional completion of a detailed risk assessment process to cover any potential risk to the quality of treated water might have highlighted the risks present in this case sufficiently that preventive measures were taken or an alternative approach used.

The true cause of the taste complaints was only realised by Scottish Water operational staff in the early hours of 26/7/06, some nine hours after the first consumer complaint was received. Had Scottish Water more quickly realised the true source of the taste, it may have been possible to reduce the extent and impact of the incident by responding more promptly with measures such as flushing. Correspondence with health professionals and environmental health officers, as well as liaison with DWQR, suggests that communications and reporting by Scottish Water to external bodies were generally satisfactory. There is anecdotal evidence to suggest that advice provided to consumers by the Scottish Water contact centre was not always appropriate or accurate, especially during the early stages of the incident. Once alerted to the true cause of the situation, Scottish Water took prompt and adequate action in taking appropriate samples and undertaking flushing to remove the stale water from the distribution system.

DWQR has issued six recommendations to Scottish Water concerning the above points, including requesting improvements to operational procedures and the risk assessment process for planned work as well as reviewing the process by which Scottish Water makes information on water quality incidents available to its contact centre and subsequently controls the updating of such information during an incident.

Read the full DWQR report on the incident