Biogeochemistry Chemistry Ecology Hazards Human impacts Pollution

Time for an oil change: How filter feeders avoid feeding on oil

Article: Fry, B., & Anderson, L. C. (2014). Minimal incorporation of Deepwater Horizon oil by estuarine filter feeders. Marine Pollution Bulletin, 80(1-2), 282–7. doi:10.1016/j.marpolbul.2013.10.018


Image courtesy of NOAA.
Image courtesy of NOAA.
Image courtesy of Skytruth
Image courtesy of Skytruth

The Deepwater Horizon oil spill that took place in the Gulf of Mexico for nearly 3 months was one of the largest oil spills on record globally and the biggest oil spill in U.S. history.  Millions of barrels of oil leaked into the Gulf causing billions of dollars worth of damage from the environment to the economy.  Immediate effects were visible, and since the well was capped, scientific studies have been underway to test for long-term effects of the spill.

Natural microbes exist in marine environments that work to biodegrade, or break down, the hydrocarbons found in oil.  When large quantities of oil are spilled and these microbes go to work, this leads to the potential of a large carbon pool that may be available as an energy source to the marine food web.  Carbon isotopes are a tracer tool that is widely used in ecology; whereas approximately 99% of carbon naturally exists as 12C, there is around 1% that exists as 13C.  There is also radioactively produced 14C, which has a very long half-life (ca. 5,700 years) and is a useful tool in terms of radiometric dating.  Stable isotopes can be measured using a combustion process coupled with “isotope ratio mass spectrometry” (IRMS), which basically separates the isotopes of an element (e.g. Carbon, Nitrogen, Sulfur, etc…) based on their slightly different atomic weights and gives you a ratio of 13C/12C in reference to a standard using the following equation: δ13C = [(Rsample/Rstandard)-1]*1000  .  Carbon isotopes are typically referred to in ‘delta’ notation, δ or ∆, followed by a number generated by the above equation and ‰ or “per mil.”  Organisms incorporate isotope signatures into their tissues that typically reflect their environment.  For example, most terrestrial plants show a δ13C value of -28‰ versus marine phytoplankton δ13C values usually range from around -21 to -23‰.  In terms of stable isotopes, oil has a δ13C value of around -27‰ and a ∆14C value of -1000‰.  Modern day ∆14C values reflect a positive signature due to the nuclear bomb testing that took place in the 1950s, so a ∆14C value of -1000‰ reflects oils extremely old age and origin.

Fry and Anderson - May oil post

What they did

Researchers here looked at filter feeders, specifically barnacles (Balanus sp.) and mussels (Geukensia demissa), to determine if species that lived within estuarine systems (mainly marshes) of coastal Louisiana were incorporating oil into their systems.  The idea is that as filter feeders, the mussels and barnacles would potentially either directly or indirectly filter any bacteria involved in the breakdown of the oil.  The authors hypothesized that there would be greater respiration in the system due to the presence of oil and that organisms in marshes with visible oil and closest to the Gulf would reflect oil signatures.


How they did it

To do this, water samples were collected to measure for respiration, which is done by performing incubations and measuring the amount of oxygen consumed (in mmol O2 consumed per m3 per day); and samples of mussels and barnacles were collected from marshes and Bays adjacent to the Gulf of Mexico.  Collection sites were both oiled and unoiled areas to provide controls for the results and taken in the months following the oil spill.  All samples were separated into shells and muscle tissue, cleaned and dried, then homogenized to be measured for δ13C and ∆14C.

The calculations

Once the ratios (R) of 13C/12C were measured on the IRMS, a correction factor needed to be performed between shells and tissues due to different incorporation of inorganic carbon.  This was done as follows:

13ε = [(Rshell/Rtissue) – 1] * 1000

It was estimated that samples reflecting no oil would show 13ε calues of 18‰ and samples with a 100% oil diet would show  13ε values of 22-25‰.  From here, they were able to estimate a percentage of oil incorporated into the organisms based off of the δ13ε and ∆14C values using the following:

% oil = 100 * (13εcontrol – 13εsample)/(13εcontrol – 13εoil)


% oil = 100 * (∆14Ccontrol – ∆14Csample)/(∆14Ccontrol – (-1000))



What they foundFry and Anderson - May oil post figures

Results were not as initially hypothesized.  The Bay sampled did not show evidence of increased respiration as expected.  The δ13ε values of both mussels and barnacles did not reflect any significant shifts indicative of oil incorporation, and the ∆14C results confirmed that there was < 1% of oil incorporated into the filter feeders.  It was surprising that so little oil appeared incorporated into these organisms when visible oil was seen on the marsh shores, but the authors discuss various reasons as to why this might be and came to the overall conclusion that it is likely the oil effects are most evident in the benthic environment versus the planktonic food web they investigated; and perhaps the oil-degraded bacteria is just not an important carbon source for these filter feeders, instead it may be lost to the atmospheric CO2 pools vs. aquatic CO2 pools.


Even though there may be little evidence of oil uptake in these filter feeding species, that does not mean there are little effects.  For examples, polycyclic aromatic hydrocarbons, or PAHs, are toxic compounds found in oil that can have lethal effects at low-dose concentrations and many fishing industries were witness to devastating effects that have still not fully recovered.  The authors propose that stronger food web effects may be seen in the deep sea region closer to the spill.

Leave a Reply

Your email address will not be published.