BP Gulf Oil Spill Response Plan Review

The gulf oil spill is now headed into its eighth week. Many have commented on the inaccuracies and weaknesses of the BP Oil Spill Response Plan, but few have analyzed it for what needs to be done to ensure the same results do not happen again. The plan is inadequate as it exists. BP’s Plan and all other plans like it need to be changed, and the state and federal agencies need to take this opportunity to collect additional information on performance of spill personnel and equipment to assist in the analysis of what is necessary to make future plans more effective.

The BP Oil Spill Response Plan is primarily an outline of the Incident Command structure, the responsibilities of the different members of the Incident Command Team, a listing of State and Federal Agencies that may need to be contacted in the event of a spill, incident forms, flow charts, decision diagrams, response organizations and equipment lists. It sounds like and is more of a resource manual that a response plan. It provides little assistance to the Incident Commander in determining what to expect for response times and effectiveness for deployed equipment or reaction of the oil to the environment.

The models upon which the spill plan is based were so inaccurate that they provided the Incident Commander with little or no understanding of when and where to expect the oil. They were woefully inadequate in assisting the Incident Commander in understanding the volume of the spill, the trajectory of where to look for the spill or the speed of the spill through the environment.

The models should be redesigned to provide more accurate projections now that there is actual data to compare. If the models are inconsistent with what happened, the models are wrong and they need to be changed to reflect reality. Even if the models are changed, they will still be wrong the next time, but they may not so grossly underestimate the speed and trajectory of the oil as to be useless to the Incident Command Team.

Only three pages of the 582 page spill plan are dedicated to a written analysis of the worst case scenario and much of that is merely an explanation of the scenario and a listing of the different response methods and their capacities to recover oil. An Incident Commander would find no assistance in this section of the plan on how to respond to the spill or what to expect from a spill of this magnitude.

There is little anywhere in the spill plan in the way of analysis to assist the Incident Commander in understanding what to expect from the response options he is provided. Equipment lists have nameplate capacities but no analysis of their effectiveness in the environment. Skimmers with nameplate capacities totaling 491,721 barrels per day are identified in the plan and give a false sense that they should be able to easily capture a spill of 250,000 barrels and remove it from the environment. But oil does not collect in one place waiting to be picked up, and nameplate capacities are not the same as effective rates.

Spill plan worst case scenarios should have estimations of how much oil is expected to escape into the environment; estimations of daily rates and how long will it take to contain and stop the spill from continuing; estimations of how much is being dispersed into the water column and how much is suspended in plumes in the water column. Of the oil collecting on the surface, the Incident Commander should have an understanding of how much will probably aggregate in sufficient quantities to be retrieved by mechanical skimmers and what percentage can be attacked with dispersants or burned? The Incident Commander should understand what percentage is escaping, the quantity of oil, the quality of the oil and when and where to expect the oil onshore.

BP needs to collect information regarding effective rates for deployment of its spill response equipment. For each piece of equipment deployed the following information should be collected:
1) the time required for equipment to be deployed onsite should be recorded, from the time it was requested to the time it was deployed onsite;
2) data should be collected regarding standby time for each piece of equipment, transportation time to get equipment to a new location for deployment, and time deployed in response to the spill;
3) during the time deployed the volume of liquid retrieved should be recorded, and data regarding the oil/water ratio should also be captured.

If this information is captured and analyzed it would provide the data necessary to calculate an effective response rate for each piece of equipment deployed. Effective response rates provide the Incident Commander with a more accurate tool than nameplate capacities in planning a response to a major oil spill.

Once the information has been collected from the spill it should be analyzed with the response team, state and federal agencies, and other stakeholders to determine the requirements for future offshore exploration and operations spill plans.

In addition, a plan that is only required to be updated once every two years is going to have a certain number of errors. I recommend annual updates and a certification by the person signing the plan as to the accuracy of the plan. That would require someone in the organization to attempt to verify all the numbers in the plan at least once a year.

Half of BP’s Incident Command Organization Chart did not have individuals identified to fill the incident command positions. I recommend that all leadership positions in the Incident Command structure should be identified in the organization chart and those individuals should be required to participate in a response drill at least annually.

If the above recommendations are followed, they won’t prevent incidents from occurring, but incident responders will have more effective tools to assist them in their response and they will be more prepared to respond to the incident.

Steve Porter is a former exploration permitting manager for a major oil company and was responsible for the preparation and approval of offshore oil spill response plans.