NOAA's Response and Restoration Blog

An inside look at the science of cleaning up and fixing the mess of marine pollution


As Oil Sands Production Rises, What Should We Expect at Diluted Bitumen (Dilbit) Spills?

Pipeline dug up for an oil spill cleanup next to a creek.

This area is where the Enbridge pipeline leaked nearly a million gallons of diluted bitumen (dilbit), a tar sands oil product, into wetlands, Talmadge Creek, and roughly 40 miles of Michigan’s Kalamazoo River in 2010. (U.S. Environmental Protection Agency)

I’ve seen a lot of firsts in the past four years.

During that time, I have been investigating the environmental impacts, through the Natural Resource Damage Assessment process, of the Enbridge pipeline spill in Michigan. In late summer of 2010, a break in an underground pipeline spilled approximately 1 million gallons of diluted bitumen into a wetland, a creek, and the Kalamazoo River. Diluted bitumen (“dilbit”) is thick, heavy crude oil from the Alberta tar sands (also known as oil sands), which is mixed with a thinner type of oil (the diluent) to allow it to flow through a pipeline.

A Whole New Experience

This was my first and NOAA’s first major experience with damage assessment for a dilbit spill, and was also a first for nearly everyone working on the cleanup and damage assessment. Dilbit production and shipping is increasing. As a result, NOAA and our colleagues in the field of spill response and damage assessment are interested in learning more about dilbit:

  • How does it behave when spilled into rivers or the ocean?
  • What kinds of effects does it have on animals, plants, and habitats?
  • Is it similar to other types of oil we’re more familiar with, or does it have unique properties?

While it’s just one case study, the Enbridge oil spill can help us answer some of those questions. My NOAA colleague Robert Haddad and I recently presented a scientific paper on this case study at Environment Canada’s Arctic and Marine Oil Spill Program conference.

In addition, the Canadian government and oil pipeline industry researchers Witt O’Brien’s, Polaris, and Western Canada Marine Response Corporation [PDF] and SL Ross [PDF] have been studying dilbit behavior as background research related to several proposed dilbit pipeline projects in the United States and Canada. Those experiments, along with the Enbridge spill case study, currently make up the state of the science on dilbit behavior and ecological impacts.

How Is Diluted Bitumen Different from Other Heavy Oils?

Dilbit is in the range of other dense, heavy oils, with a density of 920 to 940 kg/m3, which is close to the density of freshwater (1,000 kg/m3). (In general when something is denser than water, it will sink. If it is less dense, it will float.) Many experts have analyzed the behavior of heavy oils in the environment and observed that if oil sinks below the surface of the water, it becomes much harder to detect and recover. One example of how difficult this can be comes from a barge spill in the Gulf of Mexico, which left thick oil coating the bottom of the ocean.

What makes dilbit different from many other heavy oils, though, is that it includes diluent. Dilbit is composed of about 70 percent bitumen, consisting of very large, heavy molecules, and 30 percent diluent, consisting of very small, light molecules, which can evaporate much more easily than heavy ones. Other heavy oils typically have almost no light components at all. Therefore, we would expect evaporation to occur differently for dilbit compared to other heavy oils.

Environment Canada confirmed this to be the case. About four to five times as much of the dilbits evaporated compared to intermediate fuel oil (a heavy oil with no diluent), and the evaporation occurred much faster for dilbit than for intermediate fuel oil in their study. Evaporation transports toxic components of the dilbit into the air, creating a short-term exposure hazard for spill responders and assessment scientists at the site of the spill, which was the case at the 2010 Enbridge spill.

Graph of evaporation rates over time of two diluted bitumen oils and another heavy oil.

An Environment Canada study found that two types of diluted bitumen (dilbits), Access Western Blend (AWB) and Cold Lake Blend (CLB), evaporated more quickly and to a greater extent than intermediate fuel oil (IFO). The two dilbits are shown on the left and the conventional heavy oil, IFO, on the right. (Environment Canada)

Since the light molecules evaporate after dilbit spills, the leftover residue is even denser than what was spilled initially. Environment Canada, Witt O’Brien’s/Polaris/WCMRC, and SL Ross measured the increase in dilbit density over time as it weathered, finding dilbit density increased over time and eventually reached approximately the same density as freshwater.

These studies also found most of the increase in density takes place in the first day or two. What this tells us is that the early hours and days of a dilbit spill are extremely important, and there is only a short window of time before the oil becomes heavier and may become harder to clean up as it sinks below the water surface.

Unfortunately, there can be substantial confusion in the early hours and days of a spill. Was the spilled material dilbit or conventional heavy crude oil? Universal definitions do not exist for these oil product categories. Different entities sometimes categorize the same products differently. Because of these discrepancies, spill responders and scientists evaluating environmental impacts may get conflicting or hard-to-interpret information in the first few days following a spill.

Lessons from the Enbridge Oil Spill

Initially at the Enbridge oil spill, responders used traditional methods to clean up oil floating on the river’s surface, such as booms, skimmers, and vacuum equipment (see statistics on recovered oil in EPA’s Situation Reports [PDF]).

After responders discovered the dilbit had sunk to the sediment at the river’s bottom, they developed a variety of tactics to collect the oil: spraying the sediments with water, dragging chains through the sediments, agitating sediments by hand with a rake, and driving back and forth with a tracked vehicle to stir up the sediments and release oil trapped in the mud.

These tactics resulted in submerged oil working its way back up to the water surface, where it could then be collected using sorbent materials to mop up the oily sheen.

While these tactics removed some oil from the environment, they might also cause collateral damage, so the Natural Resource Damage Assessment trustees assessed impacts from the cleanup tactics as well as from the oil itself. This case is still ongoing, and trustees’ assessment of those impacts will be described in a Damage Assessment and Restoration Plan after the assessment is complete.

A hand holds a crushed mussel.

A freshwater mussel found crushed in an area of the Kalamazoo River with heavy cleanup traffic following the 2010 Enbridge oil spill. (Enbridge Natural Resource Damage Assessment Trustee Council)

For now, we can learn from the Enbridge spill and help predict some potential environmental impacts of future dilbit spills. We can predict that dilbit will weather (undergo physical and chemical changes) rapidly, becoming very dense and possibly sinking in a matter of days. If the dilbit reaches the sediment bed, it can be very difficult to get it out, and bringing in responders and heavy equipment to recover the oil from the sediments can injure the plants and animals living there.

To plan the cleanup and response and predict the impacts of future dilbit spills, we need more information on dilbit toxicity and on how quickly plants and animals can recover from disturbance. Knowing this information will help us balance the potential impacts of cleanup with the short- and long-term effects of leaving the sunken dilbit in place.


Déjà vu on the Sheboygan River: Transitioning from Cleanup to Restoration in Wisconsin

Looking upstream on the Sheboygan River from the Pennsylvania Avenue Bridge in downtown Sheboygan, Wisconsin. This section of the river was dredged in 2011 to remove sediment contaminated with PCBs and PAHs.

Looking upstream on the Sheboygan River from the Pennsylvania Avenue Bridge in downtown Sheboygan, Wisconsin. This section of the river was dredged in 2011 to remove sediment contaminated with PCBs and PAHs. (NOAA/Jessica Winter)

One of my first introductions to the problems of environmental contamination was Wisconsin’s Sheboygan River. It empties into Lake Michigan, a rich recreational, commercial, and ecological area, but unfortunately, the Sheboygan has suffered from a past filled with toxic chemicals. As an intern in the U.S. Environmental Protection Agency’s Great Lakes National Program Office in 2006, I visited this scenic river in eastern Wisconsin to learn about the techniques used for cleaning up the river’s contaminated sediments. At the time, I didn’t know that I would return with NOAA’s Office of Response and Restoration to work on the restorative process that follows cleanup: natural resource damage assessment.

A Superfund Site in the Making

Throughout the 20th century, industrial facilities released the hazardous chemicals polychlorinated biphenyls (PCBs), polycyclic aromatic hydrocarbons (PAHs), metals, and more into the Sheboygan River and adjacent floodplains. These chemicals have been measured at high concentrations in the river sediments and fish, limiting the public’s ability to use and enjoy the Sheboygan River for years. For example, resident fish and waterfowl from the river are unsafe to eat because the high contaminant levels exceed U.S. Department of Agriculture standards. To address this contamination, the EPA’s Superfund Division has designated the lower 14 miles of the Sheboygan River and the adjacent floodplains for cleanup.

On my most recent visit to the river in the fall of 2012, cleanup crews were in their final season of work on a project that has been underway for many years, beginning with emergency sediment removal in 1978. But how do you actually “clean” a polluted river like the Sheboygan?

"Geotubes," show here filled with sediment, were used to remove contaminants from Sheboygan river sediments. In the background, pipes collected weepwater which oozed out of the geotubes and left behind contaminated sediments. (U.S. Environmental Protection Agency)

“Geotubes,” show here filled with sediment, were used to remove contaminants from Sheboygan river sediments. In the background, pipes collected weepwater which oozed out of the geotubes and left behind contaminated sediments. (U.S. Environmental Protection Agency)

For the upstream stretch of the river, completed in 2006–2007, a crew had to suck up contaminated sediments from the riverbed, suspend them in water so they flow as slurry, and then pump the slurry through a pipeline. Next, they pumped it into “geotubes,” large porous bags that allow the river water to seep out but keep the sediment and solid pollutants inside. A wastewater treatment plant removed any remaining contamination from the water. Once the sediment was dry enough, it was transported to a specially designed hazardous waste landfill. Cleanup in the downstream stretch of the river in 2011–2012 used similar methods, as well as an excavator to scoop up some of the sediments and embedded pollutants.

Gearing up for Restoration

As this cleanup was winding down, my NOAA colleagues and I traveled to Sheboygan, Wis., to meet with other federal and state scientists studying the affected area. NOAA, the U.S. Fish and Wildlife Service, and the Wisconsin Department of Natural Resources serve as trustees for the public while conducting a Natural Resource Damage Assessment (NRDA). During this process, the trustees collect and evaluate data to identify the natural resources that have been injured by contamination and to quantify the resulting injuries to the environment. For example, injuries might include increased tumor rates in fish or reduced prey available for fish to eat. Luckily for us, the Sheboygan River is well-studied; we have data investigating animal populations and habitat quality from the 1970s to the present.

Fish consumption advisories, as seen posted here along the river, have been in place on the Sheboygan River since 1979.

Fish consumption advisories, as seen posted here along the river, have been in place on the Sheboygan River since 1979. (Wisconsin Department of Natural Resources/Vic Pappas)

Once the trustees know precisely what the injuries are from this pollution, they work with the public to choose projects that will address those injuries. For example, this might include creating or enhancing wetlands that will provide better areas for fish to find food. Trustees then require the parties responsible for the contamination either to fund or implement these restoration projects themselves.

In 2012, this restoration process kicked off when the trustees undertook a preliminary assessment. They examined the current state of scientific information on the Sheboygan River’s sediments, soils, water, invertebrates, fish, birds, mammals, and reptiles to determine whether it is reasonable to pursue a full damage assessment, which would compensate the public for the natural resources hurt by the Sheboygan’s history of toxic chemicals. The preassessment screen [PDF] documents this work.

What did they conclude after the preliminary assessment? That injury to these resources was likely and that damage assessment is warranted. Next, the trustees will develop an Assessment Plan that will describe the methods that will be used to quantify damages. Trustees will invite the public to comment on the Assessment Plan. Stay tuned and check out the links below to access data and documents related to this site.


  • Query Manager database: This is the general informational page for Query Manager, NOAA’s database and query tool for environmental chemistry data. Follow the link to the download page to obtain the database, map, and dictionary for Great Lakes data (which includes Sheboygan River and Harbor data) and to obtain the Query Manager software for interacting with the database.
  • NOAA is developing a new interface for accessing this data which will be available at Project DIVER is currently a work in progress.



Investigating Environmental Impacts: Oil on the Kalamazoo River

Posted sign closing river activity due to oil spill response.

The Kalamazoo River has been closed to the public since the spill in 2010. We’re examining how this has affected public recreation and tribal cultural uses. (Terry Heatlie, NOAA)

In late summer of 2010, while the nation was fixated on the massive oil spill in the Gulf of Mexico, an underground pipeline in Michigan also began gushing oil. My job has been to help investigate the environmental damage that spill caused when the oil flowed into the Kalamazoo River.

The Situation
More than 800,000 gallons of crude oil** poured out of the leaking pipeline before it was eventually shut off. It oozed through the soft, wet ground just outside of Marshall, Mich., before washing into the Kalamazoo River, one of the largest rivers in southern Michigan.

I was at a meeting in Milwaukee with my suitcase full of sandals and skirts — not exactly dressed for an oil spill — when I got called to the scene. I drove nearly nonstop to Marshall, with only a quick detour in Indiana to buy steel-toed boots and work pants.

The Challenges
When I arrived, the other scientists and I made plans to collect data on the oil’s damage. Heavy rains had caused the river to flood over its banks, and as the oil flowed approximately forty miles* down the Kalamazoo, it was also carried up onto the banks and into trees. As the flood waters receded, oil was left on overhanging branches and in floodplains.

As the flood water receded, oil was left behind on river vegetation and overhanging tree branches, as well as in yards and forested floodplains. Yellow containment boom is in the foreground. (Gene Suuppi, State of Michigan)

The river’s floodplains, full of forests and wetlands, are also home to sensitive seasonal ponds, which provide valuable habitat for fish and macroinvertebrates (aquatic “bugs” at the base of the food chain). Therefore, we needed to find out: how far did the oil make it into the floodplain, what did it contact while there, and how much oil was left?

The smell of oil was sickeningly strong at first. Residents evacuated the houses nearest to the leak, and workers within half a mile of the pipeline break had to wear respirators to protect them from inhaling fumes. Even a dozen miles downstream, I could smell the oil and feel the fumes irritating my eyes. These fumes were the light components of the oil evaporating into the air. The heavy components of the oil were left behind on the banks or gradually sank to the bottom of the river.

The sunken oil has proven difficult to clean up. This winter, spill responders have been working to quantify how much sunken oil is left and to develop and test techniques for cleaning it up.

The Science
Along with my team from NOAA’s Office of Response and Restoration, the U. S. Fish and Wildlife Service, the State of Michigan, and the Huron Band and Gun Lake Tribe of the Potawatomi joined together as trustees to assess damages that the spill caused to natural resources.

We’ve conducted a variety of studies to collect information on the impacts of the spill and repeated some of the studies to see how the environment is recovering. Now we’re gathering all this data for the official damage assessment. We’ve examined samples of fish, mussels, water, and sediments for evidence of oil-related chemicals. We’ve collected observations of oiled vegetation and records of the number and condition of animals brought to the wildlife rehab center.

Talmadge Creek cleanup crews on Aug 6, 2010.

Cleanup crews place absorbent pads to sop up oil at Talmadge Creek, near the source of the spill, on Aug 6, 2010. We also take into account the effect cleanup has on the environment. (Chuck Getter)

Unfortunately, cleanup-related activities have an environmental impact too. For example, extra boat traffic on the river during cleanup led to some riverbank erosion and crushed freshwater mussels. Our studies include these factors too. We’ll also look into the effect the spill had on public recreation (the river has been closed to the public since the spill) and on tribal cultural uses.

What Next?
We and the other trustees will seek out restoration projects that address the impacts caused by the spill, being careful to balance the projects with the results of our studies. We’ll take project ideas from the public and from watershed organizations to make sure that we choose projects that fit in well with other restoration work being done across the broader Kalamazoo River watershed.

Enbridge Energy, as the owner of the pipeline, will have the option to implement the projects themselves with oversight from us trustees, or could pay for the cost of these projects as part of a larger legal settlement.

Stay tuned and we’ll keep you updated as this story unfolds.

*Correction: This originally stated that the oil flowed thirty miles down the Kalamazoo River.

**This was later discovered to be an oil sands (or tar sands) product.