Investigations

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

Report on drinking water quality incident at:
Tullich Water Treatment Works in September 2007

The first customer contact reporting organisms in the water supply was received at 13:25
on Thursday 6 September. It was agreed with the customer to attend on the following
day. A further contact was received later that day at 20:52. The customer requested a
visit on the following day when she would be at home.

The first visit was carried out at 08:15 on 7 September. During this visit, Scottish Water
staff confirmed presence of organisms in the water supply and actions were immediately
put in place to carry out an inspection of Tullich Water Treatment Works (WTW).

The examination of the works established that organisms were present in all four filters
but particularly filter 3. Filter 3 was isolated at this point and the Treatment Operator
shut down the works and started backwash procedures with the remaining three filters.

Our Quality and Performance Team Leader attended site on 7 September. On his
advice, backwash cycles were reduced to 12hrs for all four filters. Air scouring during the
back wash cycle was increased from 1 minute to 2 minutes.

In the distribution system, flushing started at the locations of the customer complaints
from 7 September and aggressive overnight flushing was carried out on 10 September
and 11 September. The flushing programme was continued until reported sightings of
organisms dropped to levels where it was evident that supplies were returning to normal.

An increased volume of calls was received between 7 and 10 September during which
time bottled water was made available to customers. It was evident at an early stage in
the event that the issue was confined to two District Metered Areas (DMA’s) supplied
direct from the trunk main immediately downstream of the WTW (Soroba South and
Soroba North).

Visual inspections were carried out at the two service reservoirs in the distribution system
downstream of the affected DMA’s and samples taken. There was no evidence of any
organisms in the service reservoirs. Both clear water tanks at the treatment works were
drained and cleaned between 8 and 11 September. Again, there was no evidence of
organisms in the clear water tanks when these tanks were cleaned.

Samples were taken at the various stages of the Works and in the distribution system
and sent to our laboratory on 7 September 2007. From analysis of the samples, the
organisms were identified as chironomidae larvae (common name bloodworm).
Investigations into the source of the larvae have proved inconclusive.

Tullich WTW is sampled daily for microbiological parameters and all samples taken
during and since the recent problem of larvae in the water have been compliant.

In addition to monitoring customer contacts and visiting every customer who contacted
us with a complaint of organisms in the water, we also carried out random sampling at
customers’ premises within the affected area on 14 September and again on 17
September.

It was clear from our on site monitoring and the reduction in customer contacts that
supplies were returning to normal by 12 September. By 17 September, our site survey
and the low number of customer contacts gave us confidence that supplies had returned
to normal.

DWQR Assessment on the incident

This incident caused considerable alarm to consumers in the Oban area and although it was an unusual occurrence, DWQR expects Scottish Water to take all possible steps to ensure there is no repetition. Having investigated the incident, DWQR is unable to identify the source of the Chironomid larvae other than confirm that it is likely to have been one of those proposed by Scottish Water. Similarly there appear to have been no major deficiencies in Scottish Water’s operational practice that would have caused the incident. Scottish Water appear to have responded promptly to initial reports of a problem by increasing the frequency of filter washing. The localised nature of consumer complaints and the speed with which the increased wash frequencies brought the situation under control suggests that the outbreak was limited in severity and not an extensive infestation.

Scottish Water’s report on the incident contains 10 recommendations, some of which have already been completed. Key recommendations include: a review of filter operation at the site including backwash frequencies and media condition, an assessment of the benefit of installing electric insect control equipment in the filter hall, and regular sample checks on the water in the distribution system.

The design of the process at Tullich WTW is comparatively unusual, in that the filter media is entirely GAC rather than the more usual sand. GAC has been used at Tullich for a number of years following taste and odour complaints which it appears to have resolved. A possible drawback of using GAC in this way however, is that it tends to be more biologically active than sand, especially when it follows an ozonation stage as in this situation. This could provide a more favourable environment for Chironomid larvae because there is more likely to be a readily available food source. GAC is a comparatively soft filter medium, meaning frequent vigorous washing can lead to media loss and particle attrition. Industry guidance for washing GAC filters used as a secondary “polishing” stage suggests that 2-3 days is the norm. Normal operation at Tullich prior to the incident was slightly longer than this at 84 hours, and this may have been a factor, especially considering these are primary filters and likely to require more frequent washing. Consequently, it may be prudent to use shorter backwash intervals at the site.

In this respect, DWQR believes that Scottish Water’s recommendation to review filter backwash frequencies at the site is key to preventing a recurrence. The ease with which increased backwashing appears to have resolved the problem suggests this to be the case. Having said that, it may be that the use of GAC as the primary filter medium becomes uneconomic at the increased backwash frequency due to loss of filter media, and Scottish Water may wish to consider alternatives, although that is a matter for them. The other recommended actions at the site will also prove beneficial.

Now that the possible presence of Chironomids at this site is known, it is vital that monitoring is in place to provide an early warning of any emerging issue. Scottish Water plans to undertake regular sampling from hydrants in the distribution system, and DWQR welcomes this. Regular sampling of treated water leaving the treatment works should also be implemented.

This incident was very much concerned with public perception rather than a specific public health concern. Scottish Water’s communications with consumers is therefore an important part of this incident investigation. Although there is some evidence of an initial reluctance on Scottish Water’s part to provide bottled water to affected consumers, Scottish Water did visit everyone who telephoned them reporting worm-like organisms in their water supply, and they continued to provide comprehensive updates, via letters from the regional manager, to those in the two areas directly affected.

DWQR is of the opinion that, whilst Scottish Water appears to have provided a very good level of consumer care and information to residents of the two areas affected, communications with the wider community could have been better. Press reports suggest that in the early days of the incident consumers in the wider Oban area were very much aware that there was an issue, yet they felt that they were receiving little or no accurate information directly from Scottish Water. This, combined with the media reports, appears to have damaged people’s trust in their water supply. DWQR and Argyll and Bute Council are of the opinion that consumers in the wider area should have been provided with the same information as those in the affected area, and that it may have been appropriate to have extended the offer of bottled water supplies to all residents in the affected area.

A proactive media release coupled with the provision of information on the Scottish Water website should have been considered at a much earlier stage in the incident. DWQR appreciates that there is a need to balance the provision of information with the need to avoid causing undue alarm, and that this can be a difficult balance to strike, however on this occasion Scottish Water appears to have got the balance wrong. Scottish Water recognise in their report on the incident that information should have been circulated earlier to all consumers in the area and to elected council officials. Argyll and Bute council has commented that external agencies, such as Environmental Health and the NHS Board would like Scottish Water’s communications with them to have included more information on Scottish Water’s investigations and work to resolve the problem rather than focussing on the number of consumer complaints received.

The occurrence of Chironomid larvae in the water supply, whilst rare and of no consequence to public health, is significant in terms of the level of concern it generates in consumers and the loss of confidence it creates in the water supply. For this reason, the Environmental Health Officer for Argyll and Bute has requested that the Scottish Waterborne Hazard Plan be updated to include guidance on communications and the issue of alternative supplies in circumstances such as this. DWQR fully supports this suggestion as it should enable Scottish Water to react more effectively to consumer concerns if a similar incident arises again.

> read the full DWQR report on this incident