The three basic approaches for removing the oil from the water were burning the oil, filtering offshore, and collecting for later processing. On 28 April, the US Coast Guard announced plans to corral and burn off up to 1,000 barrels (160 m3) of oil each day. It tested how much environmental damage a small, controlled burn of 100 barrels (16 m3) did to surrounding wetlands, but could not proceed with an open ocean burn due to poor conditions.
BP stated that more than 215,000 barrels (34,200 m3) of oil-water mix had been recovered by 25 May. In mid June, BP ordered 32 machines that separate oil and water with each machine capable of extracting up to 2,000 barrels (320 m3) per day, BP agreed to use the technology after testing machines for one week. By 28 June, BP had successfully removed 890,000 barrels (141,000 m3) of oily liquid and burned about 314,000 barrels (49,900 m3) of oil.
In November the EPA reported that there were successful attempts made to contain the environmental impact of the oil spill, in which the Unified Command used the "situ burning" method to burn off the oil in controlled environments on the surface of the ocean to try to limit the environmental damages on the ocean as well as the shorelines. 411 controlled burn events took place, of which 410 could be quantified. Burning off approximately 9,300,000 to 13,100,000 US gallons (35,000 to 50,000 m3) on the ocean surface.
The Environmental Protection Agency prohibited the use of skimmers that left more than 15 parts per million of oil in the water. Many large-scale skimmers were therefore unable to be used in the cleanup because they exceed this limit. An urban myth developed that the U.S. government declined the offers because of the requirements of the Jones Act. This proved untrue and many foreign assets deployed to aid in cleanup efforts. The Taiwanese supertanker A Whale, recently retrofitted as a skimmer, was tested in early July but failed to collect a significant amount of oil. According to Bob Grantham, a spokesman for shipowner TMT, this was due to BP's use of chemical dispersants. The Coast Guard said 33,000,000 US gallons (120,000 m3) of tainted water had been recovered, with 5,000,000 US gallons (19,000 m3) of that consisting of oil. An estimated 11,000,000 US gallons (42,000 m3) of oil were burned. BP said 826,000 barrels (131,300 m3) had been recovered or flared. The National Oceanic and Atmospheric Administration (NOAA) estimated that about 25% of the oil had been removed from the Gulf. The table below presents the NOAA estimates based on an estimated release of 4,900,000 barrels (780,000 m3) of oil (the category "chemically dispersed" includes dispersal at the surface and at the wellhead; "naturally dispersed" was mostly at the wellhead; "residual" is the oil remaining as surface sheen, floating tarballs, and oil washed ashore or buried in sediment). However, there is plus or minus 10% uncertainty in the total volume of the oil spill.
Two months after these numbers were released Carol Browner, director of the White House Office of Energy and Climate Change Policy, said they were "never meant to be a precise tool" and that the data "was simply not designed to explain, or capable of explaining, the fate of the oil... oil that the budget classified as dispersed, dissolved, or evaporated is not necessarily gone".
Based on these estimates, up to 75% of the oil from BP's Gulf oil disaster still remained in the Gulf environment, according to Christopher Haney, chief scientist for Defenders of Wildlife, who called the government report's conclusions misleading. Haney said. "Terms such as 'dispersed,' 'dissolved' and 'residual' do not mean gone. That's comparable to saying the sugar dissolved in my coffee is no longer there because I can't see it. By Director Lubchenco's own acknowledgment, the oil which is out of sight is not benign. "Whether buried under beaches or settling on the ocean floor, residues from the spill will remain toxic for decades."
Appearing before Congress, Bill Lehr, a senior scientist at NOAA's Office of Response and Restoration, defended a report written by the National Incident Command (NIC) on the fate of the oil. This report relied on numbers generated by government and non-government oil spill experts, using an Oil Budget Calculator (OBC) developed for this spill. Based upon the OBC, Lehr said 6% was burned and 4% was skimmed but he could not be confident of numbers for the amount collected from beaches. As seen in the table above, he pointed out that much of the oil has evaporated or been dispersed or dissolved into the water column. Under questioning from congressman Ed Markey, Lehr agreed that the report said the amount of oil that went into the Gulf was 4.1m barrels, noting that 800,000 barrels (130,000 m3) were siphoned off directly from the well.
NOAA has been criticized by some independent scientists and Congress for the report's conclusions and for failing to explain how the scientists arrived at the calculations detailed in the table above. A formally peer-reviewed report documenting the OBC was scheduled for release in early October. Markey told Lehr the NIC report had given the public a false sense of confidence. "You shouldn't have released it until you knew it was right," he said. Ian MacDonald, an ocean scientist at Florida State University, claims the NIC report "was not science". He accused the White House of making "sweeping and largely unsupported" claims that three-quarters of the oil in the Gulf was gone. "I believe this report is misleading," he said. "The imprint will be there in the Gulf of Mexico for the rest of my life. It is not gone and it will not go away quickly."
By late July, two weeks after the flow of oil had stopped, oil on the surface of the Gulf had largely dissipated but concern still remained for underwater oil and ecological damage. In August, scientists had determined as much as 79% of the oil remains in the Gulf of Mexico, under the surface. In March 2011, it was reported that thousands of pounds of oil and dispersant were still collected each day from highly visible resort areas and that 17,000 lb (7,700 kg) were collected from a beach in Alabama after a winter storm.
Oil eating microbes
In August, a study of bacterial activity in the Gulf led by Terry Hazen of the Lawrence Berkeley National Laboratory, found a previously unknown bacterial species and reported in the journal Science that it was able to break down the oil without depleting oxygen levels.  Hazen’s interpretation had its skeptics. John Kessler, a chemical oceanographer at Texas A&M University says “what Hazen was measuring was a component of the entire hydrocarbon matrix,” which is a complex mix of literally thousands of different molecules. Although the few molecules described in the new paper in Science may well have degraded within weeks, Kessler says, “there are others that have much longer half-lives – on the order of years, sometimes even decades.” He noted that the missing oil has been found in the form of large oil plumes, one the size of Manhattan, which do not appear to be biodegrading very fast.
By mid-September, research showed these microbes mainly digested natural gas spewing from the wellhead – propane, ethane, and butane – rather than oil, according to a subsequent study published in the journal Science. David L. Valentine, a professor of microbial geochemistry at UC Santa Barbara, said that the oil-gobbling properties of the microbes had been grossly overstated. Methane was the most abudant hydrocarbon released during the spill. It has been suggested that vigorous deepwater bacterial bloom respired nearly all the released methane within 4 months, leaving behind a residual microbial community containing methanotrophic bacteria.
Some experts have suggested that the proliferation of the bacteria may be causing health issues for residents of the Gulf Coast. Marine toxicologist Riki Ott says that the bacteria, some of which have been genetically modified, or otherwise bio-engineered to better eat the oil, might be responsible for an outbreak of mysterious skin rashes noted by Gulf physicians.