Flip on the news and you'll see pundits a-go-go discussing the amount of oil left in the Gulf of Mexico. But how does one actually determine this amount? Diandra Leslie-Pelecky explores the fluidity of the figures and potential clean-up solutions.
If Shakespeare were alive today, he might suggest that PR/branding/marketing people should go before lawyers. The spin doctor's job is to make their company look good - even when it shouldn't. The existence of spin doctors is reason number one why we must fix our badly broken education system. If science teaches one thing, it is skepticism: People shouldn't be so anxious to be told what to think.
This week marked the first peer-reviewed study of the Gulf Coast oil spill. Big disclaimer here - it takes weeks (sometimes months or years) for research to make it from research to print, so the results are from measurements made in June; however, there are many more research reports to come. The Truth is likely to be known only after multiple research groups have published on different aspects - probably on a timescale of years - and all of their data is analyzed in aggregate. Science aims for precision, not speed.
The last few weeks have brought heated debates on radio and television arguing over exactly "how much oil is left in the Gulf". Estimates vary from "none" to "more than seventy-five percent". How can people looking at the exact same system come up with numbers that are so different?
How much oil was released into the Gulf? The National Incident Command (mostly the federal people in charge of managing the cleanup) uses a figure of about 5 million barrels and, on August 2nd, noted that about 17 percent of the oil released from the spill has been removed.
On August 19th, Ian MacDonald (a biological oceanographer from Florida State) testified in Congress and presented a very interesting fact: the 5 million barrel total includes oil that was pumped directly from the well onto ships after the first capping attempts. Initially, this larger figure doesn't make sense - why would the government use a bigger number?
Because if you include the oil pumped directly from the well to a tanker, you've "removed" a higher percentage of the oil. MacDonald (who argues that the total amount of oil that actually was released into the water was 4.1 million barrels) says that six percent was burned and four percent was skimmed. If you use his numbers, a total of 10 percent of the oil was 'removed' from the environment. The government number of 17 percent isn't wrong - but you can't compare the two numbers head to head. The higher number sounds better, but it's important for the consumer (of information) to know what that number actually represents.
A report from NOAA (National Oceanic and Atmospheric Administration) in early August estimated - and that's an important word - that 74 percent of the oil from the leak had been: captured directly from the well; skimmed; burned; dispersed chemically; dispersed naturally; evaporated; or dissolved into microscopic droplets. The rest of the oil washed ashore. On August 4th, Carol Browner said on the Today Show that "more than three-quarters of the oil is gone."
Salad dressing made with oil and vinegar (or water) presents a challenge in mixing, as I've written about in the past. Oil molecules don't like water molecules and water molecules don't like oil molecules. You need to either whisk in the oil or shake it vigorously to form an emulsion. An oil-in-water emulsion consists of very small droplets of oil suspended in water. Most salad dressing recipes include dispersants of some type. Homemade dressings rely on things like mustard and egg yolk. Commercial salad dressings use more exotic dispersants. Few people call them dispersants, but these additives encourage the formation and persistence of the emulsion, which is pretty much what a dispersant does.
I don't see any oil after shaking the bottle of vinaigrette in the fridge. Is it 'gone''? Evidently not, because the calorie count doesn't have separate columns for 'shaken' and 'not shaken'. It's definitely 'not visible', but it sure the heck isn't 'gone'.
I happen to like my vinaigrette more on the vinegar side than the oil side, so I typically use a lower ratio than the recommended 3:1 oil:vinegar ratio for a classic vinaigrette. My eating companions rarely agree with my acidic preferences, so I usually make the salad dressing with the 3:1 ratio and then add additional vinegar to a small container just for me. That decreases the oil concentration, but not the absolute amount of oil - it's just spread out more than it was before. That's not 'gone' either.
It would be really nice to have an equation that predicts how long oil takes to break down in seawater. Such an equation does exist - in fact, a number of different models exist. MacDonald's Congressional testimony makes the point that federal report (which makes no secret that its numbers/predictions are based primarily on theory) doesn't indicate which scientific formulae, models or predictors they used for their estimates.
The Science paper researchers (along with researchers at other institutions that have peer-reviewed papers coming out soon) dropped sensors into the water and some even sent a submersible to look various 'plumes'. The lead researchers on the Science paper (from Woods Hole Oceanographic Institute) reported on over 57,000 chemical analysis on a plume "only" 200 meters thick and about 2 kilometers wide over two weeks in June. The plume measured was about 3,600 feet below the surface and extended more than 20 miles southwest of the well, traveling at about four miles a day. The caveats, of course, are that this study occurred before the well was capped and two months ago.
The use of the word 'plume' may be a little misleading because this is a collection of small oil droplets, presumably formed by the dispersant or natural actions. It's not the billowing cloud you might picture. The Science paper reports significantly higher concentrations of a number of toxic compounds, including benzene at a rate of 50 micrograms per liter (which is 0.05 parts per million). One part per million is about one drop in 40 gallons. The maximum allowable concentration of benzene in drinking water is 5 parts per billion, which is 0.005 parts per million or ten times less than the concentrations they measured. (Incidentally, commercial salad dressing is required to contain 30% oil.)
Dispersing is (in my opinion) not the same thing as removing. How do we actually remove oil?
Crude oil is a collection of many different types of hydrocarbons, the exact composition of which varies from location to location. The molecules, all of which are carbon and hydrogen atoms in various configurations, are separated into fractions according to their boiling points. The lighter fractions have fewer carbon atoms (and by association, fewer hydrogen atoms) and lower boiling points. The pie chart below is representative of what might flow through a pipeline carrying Gulf Coast crude (See Appendix 6-A). Boiling points are shown in parentheses after the specific fraction name.
The light ends through medium naptha generally have a composition like gasoline, while the heavier components are more like oils. A representative molecule from the gasoline range would be octane or isooctane (both C8H18), while a representative molecule from the oils side would be paraffin wax (C25H52). The heaviest components of crude oil are like the binders for asphalt.
What everyone is counting on to actually 'remove' the oil are specific types of marine bacteria that eat oil. The bacteria with the best PR appear to be Alcanivorax borkumensis, which loves hydrocarbons for breakfast, lunch and dinner. They prefer alkanes, which are saturated hydrocarbons and very prevalent in crude oil. David Valentine of UCSB has a nice video showing the dispersant and the bacteria's effects on oil in water on the PBS Newshour website.
If you've ever watched bacteria grow on bread (either due to a science project or reluctance to clean the fridge), you know that the growth of a bacterial colony depends on specific environmental conditions. A. borkumensis likes water temperatures in the range of 20-30 C, and a 3-10 percent salt concentration, which makes it ideal in marine applications.
A group at Bangor University studies the limiting factors this bacteria's growth. We already know that the bacteria population won't grow unless there is plenty of food (not a problem in this case). The Bangor group has been studying whether you can 'fertilize' these bacteria to get them to grow faster. They are examining how nitrogen and phosphorus may stimulate growth of the bacteria to allow them to deal with very large oil spills.
These oil-gobbling bacteria need oxygen to survive. You may also have heard reports that scientists are concerned that the bacteria colonies could grow so large that they significantly de-oxygenate the water. So far, that doesn't look like a problem.
The federal report, which has been widely quoted, says that early signs suggest the oil is "biodegrading quickly." A number of scientists quickly challenged this assertion. Bacteria's metabolism decreases as the temperature decreases. Water gets colder with depth, so bacteria metabolizing oil will be less efficient at lower depths. One of the problems with dispersants is that, at some oil droplet size, the force of gravity exactly counterbalances the buoyant force, so droplets are stuck at a particular depth. It may take a very long time for these smaller droplets at lower depths to be eaten - remember that the Science paper looks at a plume 3600 feet below the surface.
There are also a lot of unknowns. We have precious little data about how the dispersants used may change the bacteria's ability to digest oil. I've been looking for hard information about exactly which components of crude oil the bacteria eat and - sort of importantly - the by-products of the eating process. Some groups have engineered bacteria so that they eat oil and produce plastic. I am suspicious that CO2 might be a by-product of the metabolism, since it's released in human metabolism of sugars (hydrocarbons with oxygen) and combustion of gasoline.
This is an interesting case study in communicating science. The federal report that came up with the 75% 'gone' number was very specific about what they meant by 'gone', even though they included some things I wouldn't necessarily include. By the time the information got reduced to a soundbite, though, the details were 'gone'.
Research scientists have been pretty unanimously vocal about their skepticism of the 75% figure. It is interesting because we usually go out of our way to emphasize the limits of validity of our conclusions, we make all kinds of caveats and generally put so many conditions on our results that our PR people want to pull out their hair (or ours). I think the vocal outcry from the scientific community is because conclusions are being presented as 'known', when in fact, they are very tenuous. There is nothing like drawing conclusions before the data are all in to get scientists riled up.
A very important role of science education - usually overlooked because it isn't easy to test - is teaching people how to think logically and rationally. This becomes even more important when people are inundated with information from all sides, including professionals whose job it is to promote specific agendas. It ought to be an attractive theme for today's students: Don't take anyone's word - prove it for yourself.
This post originally appeared on Cocktail Party Physics.
[Photograph: Daniel Beltra/Greenpeace/EPA via The Guardian]