Investigations

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

Report on drinking water quality incident at:
Ashgrove Water Treatment Works in September 2006

Scottish Water Executive Summary

Ashgrove water treatment works (WTW) supplies direct to the Kilwinning area and to Greenhead 5&6 and Greenhead 7A&B service reservoirs (SR). There is no secondary chlorination at Greenhead. Greenhead 5&6 feeds only ICI and Ardeer - mainly industrial use.

Greenhead 7A&B feeds Saltcoats and Stevenston. From mid to end of September the distribution zone from Greenhead 7A&B experienced a number of high coliform sample failures of unknown origin. Samples collected from the WTW, both SRs and the other distribution systems all passed. The residual chlorine leaving the tanks at Greenhead was found to be very low (< 0.1 mgCl/l). The chlorine residual leaving the WTW was increased from 0.3 - 1.1 (incrementally) and secondary chlorine dosing units were installed on the outlet of the SRs. This increased chlorine leaving the SRs to around 0.6 mgCl/l.

Greenhead 7A&B was taken out-of-service for cleaning and the supply fed from Greenhead 5&6. On cleaning, some debris and a small crack were found in the roof of 7A. All samples collected from 7A had passed though, so these were unlikely to have been the cause of the failures. The failures in distribution dramatically reduced after the transfer of source to Greenhead 5&6 and increased chlorine dosing, the last failure being on the 27th September.

Despite extensive investigation no definitive cause was found for the failures. A semi-stagnant main which was disturbed at the end of August may have been the source of the coliforms but direct correlation of these events is not possible. As a precaution, the valve on the main connecting the outlets of 7A/B and 5&6 has now been closed, to prevent potential disturbance of biofilm. The main recommendation is that that SR 5&6 should be taken permanently out of service and the pipework simplified. SR 7A&B could then feed this distribution system. This would increase turn-over, and, hence, residual chlorines in distribution.

DWQR Assessment on the incident

The DWQR assessment of the incident is that the persistent coliform failures in the part of Ashgrove Water Supply Zone fed from Greenhead Tank 7A/&B were, in all probability, due to some form of non-faecal contamination of the supply at or close to Greenhead tank. Scottish Water's conclusion that the cause has not conclusively been determined is correct, although their suggestion of a biofilm source arising from the complex pipework arrangement on the service reservoir site is highly feasible. DWQR concurs with Scottish Water's recommendations and action plan to prevent a recurrence; namely to simplify the pipework around Greenhead tanks, and ultimately removing Greenhead 5 and 6 from service to improve water turnover in the service reservoir. The action to obtain reliable chlorine analysis equipment is to be welcomed, although it is of concern that this was not already in place in the area prior to this incident. It is essential that all operational areas throughout Scottish Water review their provision for enabling operational staff to accurately measure chlorine residuals within the distribution system as a matter of some urgency.

Whilst endorsing Scottish Water's recommendations concerning this incident, DWQR has the following additional comments and recommendations to make, based on Scottish Water's report into the incident and DWQR's own observations of the way in which the incident was handled:

1. There were a large number of microbiological samples taken without a corresponding measurement of free and total chlorine being made. This is especially true of free chlorine residuals and the period of 17-18 September. The analysis of chlorine data is vital in the investigation of any incident of this nature, and microbiological results can be meaningless without this background information. It is hoped that the steps identified by Scottish Water in improving the facility to reliably analyse chlorine residuals in distribution will fully address this issue.
2. It was at least 20 September, one week after the first failing sample was obtained, before chlorine residuals in excess of 1 mg/l were achieved in the area affected with any degree of consistency. In the early stages of the incident, before it became clear that any contamination was unlikely to present a significant risk to public health, a more precautionary approach could have been adopted and chlorine residuals elevated more promptly, rather than repeatedly re-sampling from the area in the hope that the coliform contamination would clear of its own accord.
3. The use of accurate maps of supply zones and affected areas is essential in ensuring a systematic approach to identifying trends and possible sources of contamination. Other stakeholders, such as CPHMs and EHOs, also depend on the provision of clear and accurate information on the areas which are potentially affected. In the early stages of the incident this was not felt to be forthcoming from Scottish Water - the ideal situation would be for Scottish Water to provide maps of relevant supply zones to each local authority and health board at the start of each year, but failing this they should certainly be readily available at the start of an incident such as this one. In the latter stages of the incident, Scottish Water did improve the geographical information that it provided to external stakeholders. It should also be noted that Scottish Water cooperated fully with the relevant external stakeholders throughout the incident, and this is to be commended.
4. Some concern has been expressed that Scottish Water's distribution sample surveys placed too great a reliance on repeated re-sampling of the same properties. Whilst DWQR appreciates the merit in having certain "reference" sample locations, there is also undoubtedly a need to obtain as broad a picture as possible from across the affected area. It might also have been useful to have periodically taken a few samples from the other areas supplied by Ashgrove WTW and Greenhead tanks in order to provide the continued assurance that these areas remained unaffected, until such time as the exact source of the contamination could be positively identified.