Ignoring Quantitative Reasoning: The Initial Estimates of the Size of the Deepwater Horizon Oil Spill of 2010 and the Failure of Science in Properly Shaping the Policy Response

The initial estimate of the flow rate of now liberated crude oil following the explosion and sinking of the Deepwater Horizon oil platform turned out to be a factor of 50 times lower than the physical reality. This initial estimate, provided by the corporate owner of the oil platform, British Petroleum (BP), was a leak rate of 1,000 barrels per day (bpd). This number was not based on any scientific approach and was never put into context, for the media or the public, of whether this was a big or small number (i.e., how many bpd is equivalent to filling a bathtub for 24 h) and was simply accepted as the physical reality. As a consequence, the initial response to the disaster would plan for a scope that was much smaller than what ultimately unfolded. Furthermore, since 1,000 bpd turns out to be a small number, the initial strategy was based on the belief that the leak could be patched and therefore a fix was manageable. Here we show that (a) simple physical reasoning at the time of the occurrence would have lead to initial estimates that were close to the final estimate (determined 2 months after the initial incident) of about 50,000 bpd; (b) there was an unnecessarily slow time evolution to involve the scientific community to gather relevant data that would vastly improve the estimate and; (c) this slow evolution in unmasking the physical reality of the situation prevented a more robust governmental response to the problem. Even though the government, through National Oceanic and Atmospheric Administration (NOAA), revised the leak rate to 5,000 bpd one week after the disaster, another month would elapse before it was officially recognized that the leak rate was essentially 10 times higher.