NOAA's Response and Restoration Blog

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


1 Comment

Attempting to Answer One Question Over and Over Again: Where Will the Oil Go?

The Deepwater Horizon Oil Spill: Five Years Later

This is the first in a series of stories over the coming weeks looking at various topics related to the response, the Natural Resource Damage Assessment science, restoration efforts, and the future of the Gulf of Mexico.

Oil spills raise all sorts of scientific questions, and NOAA’s job is to help answer them.

We have a saying that each oil spill is unique, but there is one question we get after almost every spill: Where will the oil go? One of our primary scientific products during a spill is a trajectory forecast, which often takes the form of a map showing where the oil is likely to travel and which shorelines and other environmentally or culturally sensitive areas might be at risk.

Oil spill responders need to know this information to know which shorelines to protect with containment boom, or where to stage cleanup equipment, or which areas should be closed to fishing or boating during a spill.

To help predict the movement of oil, we developed the computer model GNOME to forecast the complex interactions among currents, winds, and other physical processes affecting oil’s movement in the ocean. We update this model daily with information gathered from field observations, such as those from trained observers tasked with flying over a spill to verify its often-changing location, and new forecasts for ocean currents and winds.

Modeling a Moving Target

One of the biggest challenges we’ve faced in trying to answer this question was, not surprisingly, the 2010 Deepwater Horizon oil spill. Because of the continual release of oil—tens of thousands of barrels of oil each day—over nearly three months, we had to prepare hundreds of forecasts as more oil entered the Gulf of Mexico each day, was moved by ocean currents and winds, and was weathered, or physically, biologically, or chemically changed, by the environment and response efforts. A typical forecast includes modeling the outlook of the oil’s spread over the next 24, 48, and 72 hours. This task began with the first trajectory our oceanographers issued early in the morning April 21, 2010 after being notified of the accident, and continued for the next 107 days in a row. (You can access all of the forecasts from this spill online.)

Once spilled into the marine environment, oil begins to move and spread surprisingly quickly but not necessarily in a straight line. In the open ocean, winds and currents can easily move oil 20 miles or more per day, and in the presence of strong ocean currents such as the Gulf Stream, oil and other drifting materials can travel more than 100 miles per day. Closer to the coast, tidal currents also can move and spread oil across coastal waters.

While the Deepwater Horizon drilling rig and wellhead were located only 50 miles offshore of Louisiana, it took several weeks for the slick to reach shore as shifting winds and meandering currents slowly moved the oil.

A Spill Playing on Loop

Over the duration of a typical spill, we’ll revise and reissue our forecast maps on a daily basis. These maps include our best prediction of where the oil might go and the regions of highest oil coverage, as well as what is known as a “confidence boundary.” This is a line encircling not just our best predictions for oil coverage but also a broader area on the map reflecting the full possible range in our forecasts [PDF].

Our oceanographers include this confidence boundary on the forecast maps to indicate that there is a chance that oil could be located anywhere inside its borders, depending on actual conditions for wind, weather, and currents. Why is there a range of possible locations in the oil forecasts? Well, the movement of oil is very sensitive to ocean currents and wind, and predictions of oil movement rely on accurate predictions of the currents and wind at the spill site.

In addition, sometimes the information we put into the model is based on an incomplete picture of a spill. Much of the time, the immense size of the Deepwater Horizon spill on the ocean surface meant that observations from specialists flying over the spill and even satellites couldn’t capture the full picture of where all the oil was each day.

Our inevitably inexact knowledge of the many factors informing the trajectory model introduces a certain level of expected variation in its predictions, which is the situation with many models. Forecasters attempt to assess all the possible outcomes for a given scenario, estimate the likelihood of the different possibilities, and ultimately communicate risks to the decision makers.

In the case of the Deepwater Horizon oil spill, we had the added complexity of a spill that spanned many different regions—from the deep Gulf of Mexico, where ocean circulation is dominated by the swift Loop Current, to the continental shelf and nearshore area where ocean circulation is influenced by freshwater flowing from the Mississippi River. And let’s not forget that several tropical storms and hurricanes crossed the Gulf that summer [PDF].

A big concern was that if oil got into the main loop current, it could be transported to the Florida Keys, Cuba, the Bahamas, or up the eastern coast of the United States. Fortunately (for the Florida Keys) a giant eddy formed in the Gulf of Mexico in June 2010 (nicknamed Eddy Franklin after Benjamin Franklin, who did some of the early research on the Gulf Stream). This “Eddy Franklin” created a giant circular water current that kept the oil largely contained in the Gulf of Mexico.

Some of the NOAA forecast team likened our efforts that spring and summer to the movie Groundhog Day, in which the main character is forced to relive the same day over and over again. For our team, every day involved modeling the same oil spill again and again, but with constantly changing results.  Thinking back on that intense forecasting effort brings back memories packed with emotion—and exhaustion. But mostly, we recall with pride the important role our forecast team in Seattle played in answering the question “where will the oil go?”


Leave a comment

What Does It Take to Clean up the Cleanup From an Oil Spill?

Bags of oiled waste on a beach next to a No Smoking sign.

Bags and bags of oiled waste on the beach of Prince William Sound, Alaska, following the Exxon Valdez oil spill in March 1989. (NOAA)

Imagine spilling a can of paint on your basement floor (note: I have done this more than once.). Luckily, you have some paper towels nearby, and maybe some rags or an old towel you can use to mop up the mess. When you’re finished, all of those items probably will end up in the garbage. Maybe along with some of the old clothes you had on.

You might not think much about the amount of waste you generated, but it was probably a lot more than the volume of paint you spilled—maybe even 10 times as much. That number is actually a rule of thumb for oil spill cleanup. The amount of waste generated is typically about 10 times the volume of oil spilled.

Our colleagues at the International Tanker Owners Pollution Federation (ITOPF) did a study on this very topic, looking at the oil-to-waste ratio for nearly 20 spills [PDF]. (A messy job, for sure.) ITOPF found that the general rule for estimating waste at oil spills still held true at about 10 times the amount spilled.

The Mess of a Cleanup

Cleanup workers collect oily debris in bags on the banks of the Mississippi River.

Responders collect oily debris during the M/V Westchester oil spill in the Mississippi River near Empire, Louisiana, in November 2000. (NOAA)

What kinds of wastes are we talking about? Well, there is the oil recovered itself. In many cases, this can be recycled. Then there are oily liquids. These are the result of skimming oil off of the water surface, which tends to recover a lot of water too, and this has to be processed before it can be properly disposed. Shoreline cleanup is even messier, due to the large amounts of oily sands and gravel, along with seaweed, driftwood, and other debris that can end up getting oiled and need to be removed from beaches.

Some response equipment such as hard containment booms can be cleaned and reused, but that cleaning generates oily wastes too. Then there are the many sorbent materials used to mop up oil; these sorbent pads and soft booms may not be reusable and would be sent to a landfill. Finally, don’t forget about the oil-contaminated protective clothing, plastic bags, and all of the domestic garbage generated by an army of cleanup workers at the site of a spill response.

Aiming for Less Mess

A large U.S. oil spill response will have an entire section of personnel devoted to waste management. Their job is to provide the necessary storage and waste processing facilities, figure out what can be recycled, what will need to be taken to a proper landfill or incineration facility, and how to get it all there. That includes ensuring everything is in compliance with the necessary shipping, tracking, and disposal paperwork.

The amount of waste generated is a serious matter, particularly because oil spills often can occur in remote areas. In far-off locales, proper handling and transport of wastes is often as big a challenge as cleaning up the oil. Dealing with oily wastes is even more difficult in the Arctic and remote Pacific Islands such as Samoa because of the lack of adequate landfill space. One of the common goals of a spill response is to minimize wastes and segregate materials as much as possible to reduce disposal costs.

In a 2008 article [PDF], the U.S. Coast Guard explores in more detail the various sources of waste during an oil spill response and includes suggestions for incentivizing waste reduction during a response.


Leave a comment

When Oil Spills Take You to Hawaii and the Yellowstone River in Two Days

Overview of the Yellowstone River at the site of the pipeline spill.

Overview of the Yellowstone River at the site of the pipeline spill on Jan. 19, 2015. (U.S. Environmental Protection Agency)

We get called for scientific support between 100 and 150 times a year for oil spills, chemical releases, and other marine pollution events around the nation. That averages to two or three calls per week from the U.S. Coast Guard or U.S. Environmental Protection Agency, but those calls aren’t nicely scheduled out during the week, or spread out regionally among staff in different parts of the country.

The date of an oil spill is just the starting point. Many of these pollution incidents are resolved in a day or two, but some can lead to years of work for our part of NOAA. Some oil spills make the national and regional news while others might only be a local story for the small coastal town where the spill took place.

To give you an idea, some of the incidents we worked on just last week took us from Hawaii one day to eastern Montana the next day—and we were already working on two others elsewhere. These incidents included a pipeline break and oil spill in the Yellowstone River in Montana; a mystery spill of an unknown, non-oil substance that resulted in birds stranded in San Francisco Bay, California; a tug boat sinking and releasing diesel fuel off of Oahu, Hawaii; and a fishing vessel grounded near Sitka, Alaska.

Aerial view of oil spilled along the edge of Yellowstone River.

View from an aerial survey of the spill site on the Yellowstone River, taken about six miles upstream from Glendive, Montana. (Montana Department of Environmental Quality)

The Yellowstone River spill involved a pipeline releasing oil as it ran under a frozen river. The source of the leaking oil has been secured, which means no more oil is leaking, but response operations are continuing. It is an interesting spill for several reasons. One is because the oil type, Bakken crude, is an oil that has been in the news a lot recently. More Bakken crude oil is being transported by train these days because the location of the oil fields is far from ports or existing pipelines. Several rail car accidents involving this oil have ended in explosions. Another reason the Yellowstone River spill is of particular interest is because the response has to deal with ice and snow conditions along with the usual challenges of dealing with an oil spill.

Watch footage of an aerial survey over the Yellowstone River and spilled oil:

The mystery spill in the San Francisco Bay Area is still a mystery at this point (both what it is and where it came from), but hundreds of birds are being cleaned in the meantime. The response is coordinating sampling and chemical analysis to figure out the source of the “mystery goo” coating these seabirds.

Marine diesel fuel dyed red in the ocean.

Marine diesel fuel, dyed red, is shown approximately seven miles south of Honolulu Airport on January 23, 2015. The spill came from a tugboat that sank off Barbers Point Harbor, Oahu, on January 22. (U.S. Coast Guard)

Meanwhile, the tugboat accident in Hawaii involved about 75,000 gallons of fuel oil leaking from a tugboat that sank in over 2,000 feet of water. All 11 crewmembers of the tugboat were safely rescued. We were helping forecast what was happening to the spilled oil and where it might be drifting. In addition, there was a lot of concern about endangered Hawaiian monk seals and sea turtles in the area, but no oiled wildlife have been reported.

And that brings us to the fishing vessel grounded in Alaska. At this time the vessel is still intact and hasn’t spilled any of the 700 gallons of fuel believed to be onboard. Salvors are working to refloat the vessel. Fortunately, the crew had time to cap some of the fuel tank vents before abandoning ship, which may be helping prevent oil from being released. All four crew were safely rescued.

That makes four very different spills in four very different areas … and we have to be ready to respond with oil spill models and environmental expertise for all of them at the same time. But that’s just all in a day’s work at NOAA.


4 Comments

Information about Oil Spills Is at Your Fingertips

Where and when was the biggest oil spill? How many oil spills happen each year? Is the frequency of oil spills going up or down? Where can I get information about oil spills?

We at NOAA’s Office of Response and Restoration hear these questions often, frequently from high school students looking for help writing their research papers …

We have a lot of information on our website about oil spills in general, as well as those cases in which we have provided scientific expertise during the response. In addition, we maintain IncidentNews.noaa.gov, which has information on thousands of selected historical incidents spanning 30 years of our experience responding to spills.

But NOAA only becomes involved with larger, more complex incidents in which scientific expertise is required to track or clean up spilled oil or cases with a significant threat to marine and coastal resources. Sometimes we get involved before any oil has actually spilled, such as when a ship runs aground on a coral reef and the fuel tanks have not been breached—yet. We typically respond to more than a hundred oil and chemical incidents annually, but there are thousands of smaller spills happening as well, many in marinas and urban and industrial waterways.

So what are some good sources of information on oil spills? For general statistics on oil spills in U.S. waters, I recommend that researchers go to the National Response Center website at http://www.nrc.uscg.mil.

The National Response Center (NRC) receives all reports of releases involving hazardous substances, including oil spills. Reports to the NRC activate the National Contingency Plan and the federal government’s response capabilities. The NRC maintains reports of all releases and spills in a national database going back to 1990 that you can download and search.

If you are interested in a specific pollution incident in the United States, the U.S. Coast Guard has a lot of information in their Marine Casualty and Pollution Data files. This database goes back to 1973 and captures details about marine casualty and pollution incidents that were investigated by the Coast Guard.

On the international level, the International Tanker Owners Pollution Federation (ITOPF) does a good job of providing data on oil spills from tankers and barges transporting oil. The ITOPF database goes back to 1970, and includes data from a variety of sources including maritime insurers and shipping publications.

One of the interesting trends that ITOPF data shows is that while tanker traffic has increased over the past 30 years, there has been a downward trend in oil spills originating from tankers.  In their list of the top 20 tanker accidents, 19 occurred before 1990, or pre-Exxon Valdez. Since then, many oil pollution prevention rules have been put in place and ship navigation tools have been improved. One notable example being the phase out of single-hull tankers.

Another good source of international data is CEDRE, the Centre of Documentation, Research and Experimentation on Accidental Water Pollution, based in Brittany, France. The CEDRE website has a good map and database featuring major oil spills around the world, dating back to 1917.

Speaking of oil spill data, we crunched the numbers on the locations of all of our oil spill-related responses from 2014 and came up with the following infographic:

Map of United States with numbers of oil spill responses in various coastal regions.

NOAA oil spill responses in 2014, by region. Includes actual and potential oil spills. The Gulf of Mexico, a region which produces and refines a lot of oil, also experiences the most oil spill responses NOAA is involved with of any other region. The U.S. Coast Guard in different regions takes advantage of NOAA support services in different ways, which may account for some of the very low or very high numbers of NOAA responses in various regions. (NOAA)


Leave a comment

A Final Farewell to Oil Tankers with Single Hulls

January 1, 2015 marks a major milestone in preventing oil spills. That date is the deadline which the landmark Oil Pollution Act of 1990 (OPA-90) specifies for phasing out single-hull tankers in U.S. waters. That act, passed after the 1989 Exxon Valdez oil spill in Prince William Sound, Alaska, required that all new tankers and tank-barges be built with double hulls.

Recently constructed single-hull tankers were allowed to operate, but 25 years after the Exxon Valdez, those vessels are now at the end of their operational life and will no longer be able to carry oil as cargo. The requirement was phased in gradually because of the difficultly of converting existing single-hull tankers to double hulls, and retiring the single-hull tankers more rapidly would have been a major disruption to world shipping.

Counting Down to a New Era

There won’t be a dramatic change-over on New Year’s Eve; most of the tankers calling on U.S. ports have had double hulls years before this deadline. However, one ship which was not switched over to a double hull soon enough was the tanker Athos I. This ship, carrying 13.6 million gallons of heavy crude oil, struck a submerged anchor in the Delaware River and caused a relatively large, complicated oil spill near Philadelphia, Pennsylvania, 10 year ago.

In 1992, two years after the Oil Pollution Act, the International Convention for the Prevention of Pollution from Ships (the MARPOL Convention) was amended to require all newly built tankers have double hulls. MARPOL has been ratified by 150 countries, representing over 99 percent of merchant tonnage shipped worldwide.

Stay out of Trouble by Going Double

So, what is the big issue around single vs. double-hull ships? Historically, tankers carrying oil were built with a single hull, or single shell.

While we measure oil in barrels, it is not actually shipped that way. Instead, oil is pumped into huge tanks that are part of the structure of tankers and barges. For vessels with a single hull, one plate of steel is all that separates the oil on board from the ocean. If the hull were punctured from a collision or grounding, an oil spill is pretty much guaranteed to follow. On the other hand, a ship with a double hull has two plates of steel with empty space in between them. The second hull creates a buffer zone between the ocean and the cargo of oil.

Naval architects have debated the merits of various hull designs in reducing oil spills, and using a double hull, essentially a hull within a hull, was selected as the preferred vessel design.

Close up of gash in hull on Cosco Busan cargo ship.

The cargo ship Cosco Busan lost 53,000 gallons of fuel oil when the single-hull ship hit the San Francisco-Oakland Bay Bridge in 2007. (U.S. Coast Guard)

However, the double hull requirements only apply to tankers and tank barges. Container ships, freighters, cruise ships, and other types of vessels are still built with single hulls. While these ships carry a lot less oil than a tanker, a large non-tank vessel can still carry a lot of fuel oil, and some have caused some pretty big spills, including the 2007 oil spill caused by the cargo ship Cosco Busan in San Francisco Bay.

Of course, double hulls don’t prevent all oil spills from tankers either, but the design has been credited with reducing the amount spilled, especially in the cases of low-speed groundings and collisions.

And some pretty spectacular collisions have resulted in double-hull tankers not spilling a drop.

Twenty years after the Exxon Valdez oil spill, the Norwegian tanker SKS Satilla collided with a submerged oil rig in the Gulf of Mexico. The collision tore a huge hole in the side of the oil tanker, but, thankfully, none of the 41 million gallons of crude oil it had on board was spilled.


Leave a comment

A Major Spill in Tampa Bay—21 Years Ago this Month

Two barges next to one another; one with oil spilled on its deck.

An oil soaked barge, after the 1993 Tampa Bay spill. (NOAA)

 

OR&R’s Doug Helton recalls his experience responding to a major spill in 1993.

August 10 is an anniversary of sorts.  21 years ago, I spent much of the month of August on the beaches of Pinellas County, Florida.  But not fishing and sunbathing. On August 10, 1993, three vessels, the freighter Balsa 37, the barge Ocean 255, and the barge Bouchard 155, collided near the entrance of Tampa Bay, Florida.

A barge on fire, with smoke coming form the deck.

The collision resulted in a fire on one of the barges and caused a major spill. (NOAA)

The collision resulted in a fire on one of the barges and caused a major oil spill. Over 32,000 gallons of jet fuel, diesel, and gasoline and about 330,000 gallons of heavy fuel oil spilled from the barges. Despite emergency cleanup efforts, the oil fouled 13 miles of beaches and caused injury to birds, sea turtles, mangrove habitat, seagrasses, salt marshes, shellfish beds,  as well as closing many of the waterways to fishing and boating.

The prior year I had been hired by NOAA and tasked with developing a Rapid Assessment Program (RAP) to provide a quick response capability for oil and chemical spill damage assessments, focusing on the collection of perishable data and information, photographs, and videotape in a timely manner to determine the need for a natural resource damage assessment. The emergency nature of spills requires that this type of information be collected within hours after the release. Time-sensitive data, photographs, and videotape are often critical when designing future assessment studies and initiating restoration planning—and are also used later as evidence in support of  Natural Resource Damage Assessment (NRDA) claims. The Tampa Bay spill was one of the first major responses for the RAP team.

The case was settled long ago and restoration projects have all been implemented to address the ecological and socioeconomic impacts of the spill. But some of the damage assessment approaches developed during that incident are still used today, and some of the then innovative restoration approaches are now more commonplace.

Sunset behind a bridge over a bay.

Tampa Bay, Skyway Bridge sunset, August 3, 2013. (Jeff Krause/Creative Commons)


Leave a comment

At the Trans Alaska Pipeline’s Start, Where 200 Million Barrels of Oil Begin their Journey Each Year

Man in hard hat outside at sign at start of Trans Alaska Pipeline.

NOAA’s Incident Operations Coordinator at milepost 0 of the Trans Alaska Pipeline in Deadhorse, Alaska. (NOAA)

A couple years ago I visited the southern end of the 800-mile-long Trans Alaska Pipeline in Valdez, Alaska. As the northernmost port that remains free of ice, the Valdez Marine Terminal is where crude oil from the North Slope oil fields is loaded on tankers destined for refineries on the west coast of the United States. Last month I got to visit the northern end of the pipeline in Deadhorse, Alaska, where on average 17,001 gallons of oil enter the pipeline each minute and more than 200,000,000 barrels each year [PDF].

I was in Deadhorse to meet with Alaska Clean Seas, the primary Oil Spill Response Organization (OSRO) for all of the oil exploration and production operations in Prudhoe Bay and the other nearby oil fields.

Sign in airport showing acceptable cold weather clothing for passengers.

Everyone traveling to Deadhorse, Alaska, where the Trans Alaska Pipeline begins, must follow strict Arctic fashion guidelines. (NOAA)

The flight from Anchorage was right on time, boarded quickly, and was full of jackets and hats with every industry logo in the oilfield servicing business. Safety is a big concern in a place that is so remote, and the safety policy starts at Anchorage. Nobody is allowed on the plane without appropriate clothing.

The scenery in Deadhorse is difficult to describe. It has a flat, sprawling industrial footprint surrounded by vast tundra, shallow braided rivers, and innumerable shallow ponds and lakes. All of the infrastructure is built on large gravel pads: living quarters, warehouses, huge drilling rigs, and other equipment, with multiple racks of elevated pipelines running every direction. Unheated structures sit on the ground, but heated buildings are constructed on concrete stilts to prevent thawing of the permafrost.

Deadhorse is home to the beginning of the Trans Alaska Pipeline, combining oil from five major feeder pipelines that originate in the different oil fields that comprise the North Slope. Oil takes about 15 days to get to Valdez, moving about five miles per hour. Since its construction in 1977, the Trans Alaska Pipeline System has transported nearly 17 billion barrels of oil.

While in Deadhorse, I also got to see the Beaufort Sea. Although it was close to the summer solstice (the last sunset was about a month ago), the ocean was still mostly frozen. Response boats remained staged on land, waiting for open water.

As you can gather from these descriptions and the pictures that follow, the Arctic is not a place that easily lends itself to the type and speed of oil spill cleanup possible in warmer and more accessible areas. Learn more about NOAA’s ongoing Arctic efforts in a series of reports released in April 2014.

Follow

Get every new post delivered to your Inbox.

Join 526 other followers