Reporter 438, 26 July 1999


An Appraisal of the Oil Spill that occurred at Bodington Hall on 29th March 1999


Professor Adrian McDonald


Professor Adrian McDonald

Dean, Earth and Environment Faculty

University of Leeds

Leeds LS2 9JT

Phone: 0113 2333344

Fax: 0113 2333321

Email: eedean@geog.leeds.ac.uk


Executive Summary

On the afternoon of Monday 29th March 1999 between 10,000 and 13,000 litres of red diesel fuel oil was released into Buckle Beck resulting in serious environmental damage to Buckle Beck and significant impact to Meanwood Beck.

The spill resulted from human error (The failure to indicate the correct tank for refilling. The delivery of fuel oil was to a tank already full.) that was compounded by:

  1. A standard operating procedure that was inadequate in that it did not anticipate human error and did not fail-safe.
  2. A bunding and tanking design that was flawed in that (i) access to surface water drains was located within the bunding and (ii) venting was to the exterior of the bunding.
  3. An inspection regime that was insufficient in that if failed to detect and remedy (i) the missing sealing plate to the surface drain and (ii) an insecure oil tank inspection cover.

Those University employees directly involved at Bodington site were neither reckless nor malicious. All University staff reacted to minimise the impact to the aquatic environment as soon as the spill was identified and sought to comply with all recommendations for remedial action proposed by the Environment Agency.

The following actions have taken place:

  1. All inadequacies in the design and maintenance at the boiler house at Bodington Hall have been, or are in the course of being, rectified.
  2. A new written standard operating procedure is in place.
  3. All other oil storages have been reviewed and, where necessary, modified to ensure that similar spills cannot occur.

In the longer term the University should consider conducting a full environmental review and audit of its activities and should have a senior manager designated to oversee the environmental performance of the institution.

Contents

The Brief

Causes

Standard Operating Procedure

Design Considerations

Drainage

Vent-pipes

Switchgear

Maintenance

Immediate Actions

Longer Term Considerations

Acknowledgements

 

The Brief

No written brief was provided for this report. A brief was agreed with Adrian Slater, the University Solicitor, as follows:

  1. To identify clearly the nature of the incident.
  2. To identify why this incident occurred.
  3. To propose short-term reactions.
  4. To consider long-term lessons for the environmental strategy of the University.

It was agreed that the report be written in plain English to make it accessible to a wide audience.

The report is based on witness statements, sub-contractor reports, commissioned reports from consultants, field observations and documentary evidence from Estates, Uniserv and Bodington Hall staff.

The Incident

The incident arose when a standard delivery of 19,000 litres of fuel oil was pumped into a tank that was already full. The tank is one of three tanks contained in a secure, roofed, locked, bunded storage facility. The excess oil spilt into the bund from an insecure inspection hatch on the top of oil storage tank number 1. A drain seal cover had been removed some time previously and thus the bund integrity was not maintained. The bund did not operate satisfactorily to retain the oil spill. The majority of the spill reached the Buckle Beck and thence Meanwood Beck and the River Aire (Figure 1). The delivery took place around 12.00 on Monday 29th March and was investigated by the Environmental Agency that afternoon following public complaints to the Agency. University staff were alerted by the Agency at approximately 16.00. University staff took immediate action once alerted to the incident. Senior managers reached the site within 45 minutes. Bodington staff had by this time already commissioned specialist oil recovery contractors.

The volume of the oil spill is estimated to be 16,000 litres. In making this estimate it is assumed that the difference between the tank volume, 25,000 litres, and the delivery volume, 19,000 litres, is divided equally into dead storage and head space. Thus it is assumed that the 3,000 litres of headspace would be filled by the new delivery and the residual 16,000 litres would spill to the bund. Not all of this oil spill reached the Meanwood Beck. From the boiler-house floor plans and the witness statement estimates of oil depth in the bund, 3,000 – 6,000 litres of oil were pumped from the bunding by Rotherham and Global Environmental Services. Thus the most likely volume of oil to have reached Buckle Beck and Meanwood Beck is between 10,000 and 13,000 litres. Best and worst case assumptions are shown in Figure 2.

However, blockages in the drain system (fortuitous in that it delayed the escape of oil and allowed 3,000 – 6,000 litres to be recovered) may mean that some of the oil has passed into soil storage.

Independent views on the damage caused by the University and necessary remedial action have been requested by the University from Professor Masterson of Chemex International. The Environment Agency is of course conducting its own independent assessment.

 

The key elements of damage are as follows:

  • Ecological damage to the river system is considerable. This is a significant incident. The 50% to 90% loss of invertebrate population reported by the Environment Agency over most of the Buckle and Meanwood Beck will have a significant effect on the biological well-being of the stream.

  • Some oil will remain in the sediment but, despite some consultants' reports, no attempt to remove or remediate the sediment should be undertaken other than with the specific agreement of the Environment Agency. River sediments are complex biological habitats having mature biofilms covering inorganic and organic sediments of a size distribution determined by the hydraulic characteristics of particular sites. Intervention should be a last resort.

  • Perceived and real concerns of riparian owners over the longer term impacts of the spill on the quality of the water.

  • The impact on the reputation of the University as a concerned major landowner in the area. There is no history of similar incidents nor has the University been subject to prosecution in relation to failures in environmental compliance.

 

 

Causes

Heating at Bodington Hall is supplied by oil-fired boilers which are fuelled from three, 25,000 litre storage-tanks. The tanks are contained within a common, secure, roofed bund of approximately 60m3 capacity. Oil is delivered to the boiler house by external contractors. The contractor knows which tank to fill by looking at the numbered disk displayed in the boiler house window.

The immediate cause of the oil spill was the failure to display the correct tank number for refilling. The tank to be filled was tank 2 but the disk number displayed, tank 1, remained unchanged from the previous week because the member of Bodington staff who had undertaken to order more fuel was distracted by other matters at the Halls and failed to change the disk number. This member of staff had not previously ordered fuel without the advice of a University engineer from Uniserv. It is the responsibility of the University engineer to change the disk in the window. In taking on this task without authority and outwith his normal job responsibilities, it is likely that he was unaware of the full significance of the disk indicator system and acted on impulse and in good faith to secure the heating at the Halls over the Easter period. The boundaries of responsibility between Uniserv and Bodington were unclear and undocumented.

However, the fundamental cause was the inadequacies in the (i) operation, (ii) design and (iii) maintenance of the storage facility at Bodington Hall. Nevertheless, the system had operated for over 30 years without fault.

 

Standard Operating Procedure

No written standard operating procedure existed at the Bodington site in respect of refueling. Those procedures that existed in unwritten form at the time of the spill were flawed. The procedures assumed that there would be no human error. Further, in the event of a human error the procedures did not fail-safe but were highly likely to result in a significant spillage. Fuel oil deliveries to the Bodington site were typically unsupervised. This is in contradiction to the Environment Agency Oil Care Code that specifies:

"Delivery. Supervise all delivery of oil to ensure that storage tanks are not over-filled and that the oil is delivered into the right tank."

Delivery of fuel oil was initiated by a phone call to the fuel oil contractors and confirmed by written order. Neither phone call nor written orders specified the tank number to be filled. As stated above, the sole indication of the tank to be filled was made by a disk displayed in the boiler room window. The fuel delivery driver:

  1. Cannot check fuel dials that are set in the locked boiler room.
  2. Cannot check fuel floats, which are locked in the boiler room and at the time of my inspection, were obstructed for at least half their length.
  3. Cannot see the storage tanks from the delivery bay.
  4. Cannot see the refueling disks from the filling point.

Thus if the indicator disk is incorrectly displayed, is unchanged from the previous week (as was the case in this instance) or if the driver connects to the wrong inlet a spill to the bunding will occur. Neither the driver nor Bodington staff will have any indication that a spill is occurring. There is no high level alarm on the tanks to warn of overfilling.

There was no clear written division of responsibility between Uniserv and Bodington staff. The unwritten procedures appear imprecise.

Design Considerations

The design of the oil fuel storage facility is in many respects inadequate in respect of:

  1. Positioning of drains and other pipe-works.
  2. The positioning of venting.
  3. Switchgear and other electrical facilities within the bund.

 

Drainage

Pollution prevention guidelines are issued by the Environment Agency in England and Wales and by SEPA in Scotland to provide managers with practical guidelines in the interpretation of good practice and legal requirement. PPG 2 applies in this case. In that guideline it is identified that:

"ideally, pipework should not pass through the bund wall"

and further that

"there must be no outlet directly connecting the bund to any drain, sewer or water course or discharging onto a yard or unmade ground".

In the case of the Bodington fuel storage site a rainwater downpipe is routed internally in the building to the floor of the bund. Adjacent to this on the floor of the bund is an inspection chamber that is normally sealed with a metal cover. This design arrangement appears to have existed since the construction of the boiler house some 30 years ago. It is, however, now clearly out of date and is not in accord with current pollution prevention guidelines.

 

Vent-pipes

PPG2 under Section 7C states that vent-pipes should be directed so that:

"any discharge from them (e.g. in the event of a tank being overfilled) passes into the bund."

This is not the case at Bodington. It may be, however, that any vented oil might be contained within the delivery vehicle bay and this could be ensured by minor modifications to the delivery bay drainage. In any event, had the inspection hatch been secure, oil discharging from the vent pipes would have been seen by the delivery driver and the majority of the incident averted.

Instructions on the use of valving in this bay, which were inadequate, have been upgraded since the oil spill incident.

 

Switchgear

There is a considerable amount of wiring and switchgear fixed to the bund walls. I am given to understand that this is "non-sparking" switchgear and that fuel oil is hard to ignite which I accept. Nevertheless at other sites at which I have conducted formal environmental investigations there appears to be significantly less electrical switchgear within the bund area. I advocate a review of the bund wiring.

 

Maintenance

Each of the storage tanks has an inspection plate that should be bolted into place thus sealing the tank completely other than to the vent pipe. Had this been the case on tank number 1 the additional oil would have been pumped up the vent pipe and would have discharged within sight of the driver. In all likelihood then the driver would have seen the discharge and ceased the pumping. Unfortunately at some stage the inspection cover has been loosened and not resealed. Therefore, the oil spilled from the inspection cover and drained into the bunding, a circumstance that could not be seen by the delivery driver.

The bund would have had the capacity to hold the entire oil spill and in that circumstance precisely when it would have been discovered that the bund was full of oil is a matter of conjecture. Unfortunately again, the cover on the drain inspection chamber was not in place and the oil drained through that surface-water channel and eventually to Buckle Beck. There appears to be no record or schedule of maintenance in respect of the bund. For this site the schedule of maintenance needs to be more clearly and fully specified.

The particular routing of the oil spill from the drain in the bund to Buckle Beck has been a matter of some debate since there are blockages and breaks in the surface-water drain system. This is a matter, however, of little consequence and is serving only to confuse the University’s analysis of the failures at Bodington. Whether the oil discharged with a surface drain to a river or through a foul drain to a sewage treatment works would simply have moved the pollution event to a different location.

 

Immediate Actions

The Environment Agency alerted university staff at approximately 16.00. University staff took immediate action once alerted to the incident. Senior managers reached the site within 45 minutes. Bodington staff had by this time already commissioned specialist oil recovery contractors. Two companies were immediately commissioned (i) to pump the remaining oil from the bund to tankers and thus stop the pollution at source and (ii) to pump accumulated oil from river pools. A third company was instructed to support the oil clean-up work of the Agency and was on site that evening. University staff worked in cooperation with the Environment Agency throughout and sought to satisfy any requirements identified. The University is unaware of any remedial actions required by the Agency that were not undertaken.

Within 72 hours of the oil spill a list of immediate actions was agreed between officers and staff of the University attending the incident. These actions included:

  1. The suspension of further fuel deliveries until essential remedial actions have been under taken.
  2. The sealing of the inlet gully and drain within the bund wall.
  3. The routing of the internal rainwater pipe to an external position.
  4. The supervision of all future oil deliveries by an appropriate member of Bodington/Uniserv staff.
  5. Replacement of the instruction board at the intake housing.
  6. Locking of the intake housing.
  7. Development of written operating procedure that incorporated the removal of all disks after delivery, the checking of the tank state at the time of delivery and the visual checking of the correct inlet attachment. The procedures require a member of University staff to be present during the fuel delivery.

In addition, the process of operating procedures is to be formally recorded and staff duties have been formalised in writing. The procedures must identify a back-up person or procedure in the event that the appropriate person is unavailable to attend to discharge the allocated responsibility.

The further 23 oil heating systems and farm tanks under the authority of the University have been reviewed with respect to their environmental security.

Conversion of the Bodington boilers to gas-fired operation is planned and will be implemented in summer 1999.

Longer Term Considerations

The University is a large and complex organisation. It has the infrastructure requirements of a modest town and handles a variety of materials, some of which are likely to be environmentally sensitive. In the light of the scale and complexity of the organisation and given the University’s role in the Leeds Initiative and in the environmental section of the initiative and, further, given the University’s position in the "Environment City" of Leeds, it is in my view incumbent upon the University to consider very seriously how it can best discharge its environmental stewardship.

The University should have a clear and comprehensive environmental policy. To do this the University should consider commissioning either internally or externally an environmental review and audit. However, the University will have to appreciate that environmental inadequacies identified through such a review and audit would have to be corrected and this is not without a serious financial consequence. Therefore, it is proposed that the University sets in train:

  1. a pilot study to determine the best structure through which
  2. to develop a progressive and rolling programme of environmental review. Such a programme would consider environmental policy, current performance, monitoring, reporting, improvement and public communication.

A senior member of staff should have explicit responsibility for environmental performance. Appropriate resources should be made available to enable the University to fulfill its environmental responsibilities.

 

Acknowledgements

This report has benefited from discussion with many members of the University and from the help of staff at Bodington Hall. Adrian Slater and his colleagues in the University Legal Section has spent much time gathering information from officers and members of the University who manage the infrastructure. Neil Stanley of the Department of Law made valuable comments on the drafts. Val Marrison, as always willing to work far beyond her job specification, typed this report.



[Main news stories | News in brief | Events | Notice board]


HTML by Hannah Dee