# EPA data show dispersants plus oil are more toxic than either alone

*Published:* 2010-06-14
*Author:* 

*Richard Denison, Ph.D., is a Senior Scientist.*

[In an earlier post](http://blogs.edf.org/nanotechnology/2010/06/06/another-bp-leak-%e2%80%93-this-time-its-their-2009-gulf-of-mexico-oil-spill-contingency-plan/), I noted in haste some apparent discrepancies between EPA and BP acute toxicity data on the Corexit® dispersants. Little did I realize that the data mixup was actually telling me something much more significant: that the dispersant maker’s own test data demonstrate that the combination of oil *plus* dispersant is quite a bit more toxic than the dispersant alone and – even more significant – the combination is more acutely toxic than the oil by itself.

Let me repeat that: The data indicate that dispersed oil is more toxic than undispersed oil. EPA has posted the dispersant manufacturer Nalco’s “Technical Product Bulletins” for each of the dispersants that have been used in the Gulf: [Corexit® EC9527A](http://www.epa.gov/oem/content/ncp/products/corex952.htm) and [Corexit® EC9500A](http://www.epa.gov/oem/content/ncp/products/corex950.htm).

Section VII of each of the bulletins shows the toxicity data for a) dispersant alone, b) the reference oil used in the test, No. 2 fuel oil, and c) a mixture of dispersant and test oil at a 1:10 ratio. Here are the data (remember, the lower the value, the more toxic the substance):

  **LC50\* values in parts per million (ppm)**  **Menidia (inland silverside fish)  
96-hour test****Mysidopsis  
(mysid shrimp)  
48-hour test****Corexit® EC9500A**Dispersant alone25.2014.57No. 2 fuel oil alone10.7216.12Dispersant : oil  
(1 : 10)2.613.40**Corexit® EC9527A**Dispersant alone32.2324.14No. 2 fuel oil alone10.7216.12Dispersant : oil  
(1 : 10)4.496.60\*LC50 = concentration at which 50% of the test organisms are killed within the duration of the test.To put these numbers in perspective, consider this calculation (thanks to Sayer Ji at [Gulf Oil Spill Truth](http://oilspilltruth.wordpress.com/dispersants/ "Gulf Oil Spill Truth") for starting me down this path of calculations):

Just 2.61 parts per million of a 1:10 mixture of Corexit® EC9500A and oil kills 50% of silverside fish within 96 hours, at least under the test conditions.

- Assuming uniform mixing, that means that 1 gallon of that mixture added to *383,142 gallons of water* would be sufficiently toxic to do the same.
- So far BP has added at least 1.2 *million* gallons of pure Corexit® dispersants to the Gulf. Assuming it’s mixed with 10 times as much oil, that’s enough dispersant to render equally toxic nearly 7,000,000 Olympic swimming pools worth of water. (There are about [660,000 gallons in an Olympic-sized swimming pool](http://www.arthurspools.com/above_ground_pool_information/olympic_pool_size.htm), the dimensions of which are about 164 x 82 x 6 feet.)

Assuming a water depth of 5000 feet at the spill site, that much water would occupy a block that is 10,600 x 10,600 x 5,000 feet, or ***4 square miles on the surface of the water extending all the way down to the ocean floor***. That’s the volume of water that would be sufficiently toxic to kill half of the fish residing in it within 4 days. Diluting that another 10-fold (into the water underneath 40 square miles of the Gulf) would still kill 5% of the fish, and another 10-fold (400 square miles) would still kill 0.5% of the fish.

Now obviously these volumes are a tiny fraction of the [volume of the Gulf](http://wiki.answers.com/Q/How_many_gallons_of_water_are_in_the_Gulf_of_Mexico), or even of the region of the Gulf where the dispersant have been sprayed and injected.

But that still adds up to a lot of extra toxicity to release into the Gulf, especially when you consider it could well be *increasing*, rather than decreasing the acute lethal toxicity of the oil.

And what about less lethal effects, e.g., killing only 1 of every 1000 fish … or killing more sensitive organisms … or effects from exposures longer than 96 hours … or sublethal effects, which can be expected to occur at much lower concentrations?

We know nothing about those possible effects because the testing for them hasn’t been done – unless you count this huge, uncontrolled experiment we’ve now embarked on.