The Deepwater Horizon oil spill was discovered on the afternoon of 22 April 2010 when a large oil slick began to spread at the former rig site. According to the Flow Rate Technical Group, the leak amounted to about of oil, exceeding the 1989 Exxon Valdez oil spill as the largest ever to originate in U.S.-controlled waters and the 1979 Ixtoc I oil spill as the largest spill in the Gulf of Mexico. BP has challenged this calculation saying that it is overestimated as it includes over of oil which was collected before it could enter the Gulf waters.

Spill flow rate In its permit to drill the well, BP estimated the worst-case flow at . Immediately after the explosion the USCG did not estimate any oil leaking from the sunken rig or from the well. On 24 April, USCG Rear Admiral Mary Landry announced that a damaged wellhead was indeed leaking. She stated that "the leak was a new discovery but could have begun when the offshore platform sank . . . two days after the initial explosion." Initial estimates by USCG and BP officials, based on remotely operated vehicles as well as the oil slick size, indicated the leak was as much as . Outside scientists quickly produced higher estimates, which presaged later increases in official numbers. Official estimates increased from on 29 April, to on 27 May, to on 10 June, and to between , on 15 June. Internal BP documents, released by Congress, estimated the flow could be as much as , if the blowout preventer and wellhead were removed and if restrictions were incorrectly modeled.

+Progression of oil spill flow rate estimates Source Date Barrels per day Gallons per day Cubic metres per day BP estimate of hypothetical worst-case scenario (assumes no blowout preventer) Permit 162,000 6,800,000 25,800 United States Coast Guard 23 April (after sinking) 0 0 0 BP and United States Coast Guard 24 April 1,000 42,000 160 Official estimates 29 April 1,000 to 5,000 42,000 to 210,000 790 Official estimates 27 May 12,000 to 19,000 500,000 to 800,000 1,900 to 3,000 Official estimates 10 June 25,000 to 30,000 1,100,000 to 1,300,000 4,000 to 4,800 Flow Rate Technical Group 19 June 35,000 to 60,000 1,500,000 to 2,500,000 5,600 to 9,500 Internal BP documents hypothetical worst case (assumes no blowout preventer) 20 June up to 100,000 up to 4,200,000 up to 16,000 Official estimates 2 August 62,000 2,604,000 9,857

Official estimates were provided by the Flow Rate Technical Group—scientists from the National Oceanic Atmospheric Administration (NOAA), United States Geological Survey (USGS), Bureau of Ocean Energy Management, Regulation and Enforcement (BOEMRE), U.S. Department of Energy (DOE), and outside academics, led by USGS director Marcia McNutt. The later estimates were believed to be more accurate because it was no longer necessary to measure multiple leaks, and because detailed pressure measurements and high-resolution video had become available. According to BP, estimating the oil flow was very difficult as there was no underwater metering at the wellhead and because of the natural gas in the outflow. The company had initially refused to allow scientists to perform more accurate, independent measurements, saying that it was not relevant to the response and that such efforts might distract from efforts to stem the flow. Former EPA Administrator Carol Browner and Congressman Ed Markey (D-MA) both accused BP of having a vested financial interest in downplaying the size of the leak in part due to the fine they will have to pay based on the amount of leaked oil.

The final estimate reported that were escaping from the well just before it was capped on 15 July. It is believed that the daily flow rate diminished over time, starting at about and decreasing as the reservoir of hydrocarbons feeding the gusher was gradually depleted.

In a 3 December 2010, statement, BP claimed the government overestimated the size of the spill by between 20% and 50%. A document submitted by BP to the commission, NOAA, and The Justice Department says that "they rely on incomplete or inaccurate information, rest in large part on assumptions that have not been validated, and are subject to far greater uncertainties than have been acknowledged. Representative Markey, a member of the House energy panel that is investigating the spill, said in a statement that BP has done whatever it could to avoid revealing the true flow rate of the spill. "With billions of dollars at stake, it is no surprise that they are now litigating the very numbers which they sought to impede." A BP spokesperson said that BP "fully intends to present its own estimate as soon as the information is available to get the science right."

Spill area and thickness

The oil's spread was initially increased by strong southerly winds caused by an impending cold front. By 25 April 2010, the spill covered and was only from the ecologically sensitive Chandeleur Islands. A 30 April 2010 estimate placed the total spread of the oil at . The spill quickly approached the Delta National Wildlife Refuge and Breton National Wildlife Refuge. On 19 May 2010, both NOAA and other scientists monitoring the spill with the European Space Agency Envisat radar satellite stated that oil had reached the Loop Current, which flows clockwise around the Gulf of Mexico towards Florida and then joins the Gulf Stream along the U.S. east coast. On 29 June 2010, NOAA determined that the oil slick was no longer a threat to the loop current and stopped tracking offshore oil predictions that include the loop currents region. The omission is noted prominently on the ongoing nearshore surface oil forecasts that are posted daily on NOAA's website.

On 14 May 2010, the Automated Data Inquiry for Oil Spills model indicated that about 35% of a hypothetical spill of light Louisiana crude oil released in conditions similar to those found in the Gulf would evaporate, that 50% to 60% of the oil would remain in or on the water, and the rest would be dispersed in the ocean. That means that over of oil remained in the Gulf.

The wellhead was capped on 15 July 2010, and by 30 July, the surface oil appeared to have dissipated more rapidly than expected. Some scientists believe that the rapid dissipation of the surface oil may have been due to a combination of factors that included the natural capacity of the region to break down oil (petroleum normally leaks from the ocean floor by way of thousands of natural seeps and certain bacteria can consume it.); winds from storms appeared to have aided in rapidly dispersing the oil, and the cleanup response by BP and the government helped control surface slicks. As much as 40% of the oil may have simply evaporated at the ocean surface, and an unknown amount remains below the surface.

However, many scientists dispute the report's methodology and figures. Scientists said much oil was still underwater and could not be detected. According to the NOAA report released on 4 August 2010, about half of the oil leaked into the Gulf remains on or below the Gulf's surface. Some scientists are calling the NOAA estimates "ludicrous." According to USF chemical oceanographer David Hollander, while 25% of the oil can be accounted for by burning, skimming, etc., 75% is still unaccounted for. The federal calculations are based on direct measurements for only of the oil spilled – the oil burned and skimmed. According to Bill Lehr, an author of the NOAA report, the other numbers are "educated scientific guesses," because "it is impossible to measure oil that is dispersed". FSU oceanography professor Ian MacDonald called it "a shaky report" and is unsatisfied with the thoroughness of the presentation and "sweeping assumptions" involved. John Kessler of Texas A&M, who led a National Science Foundation on-site study of the spill, said the report that 75% of the oil is gone is "just not true" and that 50% to 75% of the material that came out of the well remains in the water in a "dissolved or dispersed form". On 16 August 2010, UGA scientists said their analysis of federal estimates show that 80% of the oil the government said was gone from the Gulf of Mexico is still there. The Georgia team said "it is a misinterpretation of data to claim that oil that is dissolved is actually gone".

Oil sightings

Oil began washing up on the beaches of Gulf Islands National Seashore on 1 June 2010. By 4 June 2010, the spill had landed on of Louisiana's coast, had washed up along the Mississippi–Alabama barrier islands, and was found for the first time on a Florida barrier island at Pensacola Beach. On 9 June 2010, oil sludge began entering the Intracoastal Waterway through Perdido Pass after floating booms across the opening of the pass failed to stop the oil. On 23 June 2010, oil appeared on Pensacola Beach and in Gulf Islands National Seashore, and officials warned against swimming for east of the Alabama line. On 27 June 2010, tar balls and small areas of oil reached Gulf Park Estates, the first appearance of oil in Mississippi. Early in July 2010, tar balls reached Grand Isle, but 800 volunteers were cleaning them up. On 3 and 4 July 2010, tar balls and other isolated oil residue began washing ashore at beaches in Bolivar and Galveston, though it was believed a ship transported them there, and no further oil was found 5 July. On 5 July 2010, strings of oil were found in the Rigolets in Louisiana, and the next day tar balls reached the shore of Lake Pontchartrain. The amount of Louisiana shoreline affected by oil grew from in July to in late November 2010.

On 10 September 2010, it was reported that a new wave of oil suddenly coated of Louisiana coastline and marshes west of the Mississippi River in Plaquemines Parish. The Louisiana Department of Wildlife and Fisheries confirmed the sightings. On 23 October 2010, it was reported that miles-long stretches of weathered oil had been sighted in West Bay, Texas between Southwest Pass, the main shipping channel of the Mississippi River, and Tiger Pass near Venice, Louisiana. The sightings were confirmed by Matthew Hinton of The Times-Picayune.

At the end of October 2010, two research vessels studying the spill's effect on sea life found substantial amounts of oil on the seafloor. Kevin Yeager, a USM assistant professor of marine sciences found oil in samples dug up from the seafloor in a radius around the site of the Macondo well. The oil ranged from light degraded oil to thick raw crude. The sheer abundance of oil and its proximity to the well site, though, makes it "highly likely" that the oil is from the Macondo well. A second research team turned up traces of oil in sediment samples as well as evidence of chemical dispersants in blue crab larvae and long plumes of oxygen-depleted water emanating from the well site off Louisiana's coast.

In late November 2010, Plaquemine Parish, Louisiana coastal zone director P.J. Hahn reported that more than of oil had been sucked out of nearby marshes in the previous 10‑day period. In Barataria Bay, Louisiana, photos and firsthand accounts show oil still reaching high into the marshes, baby crabs and adult shrimp covered by crude and oil slicks on the surface of the water.

In January 2011, a spill commissioner reported that tar balls continue to wash up, oil sheen trails are seen in the wake of fishing boats, wetlands marsh grass remains fouled and dying, and crude oil lies offshore in deep water and in fine silts and sands onshore. On 26 May 2011, the Louisiana Department of Environmental Quality extended the state of emergency related to the spill. In April 2012, oil was found dotting of Louisiana's coast. As of December 2012, of coastline remain subject to evaluation and/or cleanup operations.

Underwater oil plumes