In May 2000, the US stopped deliberately blurring the GPS signal civilians use, and the White House said car navigation that struggled to tell parallel highways apart would get up to ten times more accurate overnight

On 1 May 2000, the White House announced that the US would stop deliberately degrading the GPS signal available to the public, starting at midnight. Its fact sheet gave a problem with car navigation as an example. GPS could place a car only “to within a hundred meters”, so in areas where “multiple highways run in parallel”, the degraded signal “made it difficult to determine which one the car was on”. In a statement that day, President Bill Clinton said civilian users would be able to pinpoint locations “up to ten times more accurately than they do now”.

Nobody had to buy anything. The government’s GPS site told owners that existing receivers “should be getting the higher accuracy right now without any modifications”. As the fact sheet put it, “It’s rare that someone can press a button and make something you already own worth more, but that’s exactly what’s happening today.”

What was being done to the signal

The feature had a name, Selective Availability, and GPS.gov describes it as “an intentional degradation of public GPS signals implemented for national security reasons”. A 1995 National Academies report explains that it worked in two ways. The satellites’ clock frequency was made to fluctuate, which the report calls dither, and the navigation message carried deliberately incorrect orbit data, which it calls epsilon.

Both matter because a receiver finds its position by timing signals from several satellites against those clocks and orbits. The clocks are precise enough that the system has to correct them for about 38 microseconds a day of relativistic drift. Selective Availability spoiled that precision for civilian users on purpose.

The report gives the target the degraded signal was held to: a horizontal accuracy of 100 metres for a standalone civilian receiver. The White House used the same figure for car navigation.

Every satellite at once, a few minutes after midnight

Air Force Space Command would carry out the order, Assistant Secretary of Defense Arthur Money told reporters at the White House briefing. GPS.gov later put the moment at “a few minutes past midnight EDT after the end of May 1, 2000” and says the change happened across the whole constellation simultaneously.

The government’s first-week data page puts the switch at about 04:05 UTC, with every satellite dropping the error “over an interval of a minute or two”. Reference stations from Amman, Jordan, to Casey station in Antarctica and Kokee Park in Hawaii recorded the drop as it happened.

That page also compares two full days at a reference station in Erlanger, Kentucky. On 1 May, with the error switched on, 95 percent of position fixes fell inside a circle 45.0 metres in radius. On 3 May, with it off, 95 percent fell inside 6.3 metres.

The page offers its own picture of the difference. With Selective Availability on, “you really only know if you are on the field or in the stands at that football stadium; with SA switched off, you know which yard marker you are standing on.”

The Gulf War precedent

The degradation had been turned down before, when US troops needed it gone. During the 1990–91 Persian Gulf War, the Defense Department had only a small number of military GPS receivers, and the National Academies report says this “prompted a National Command Authority decision to turn SA to zero during the war”. The department bought thousands of civilian receivers instead.

London’s Science Museum says some units had as few as two instruments for 100 vehicles, and some soldiers had relatives buy civilian sets and ship them out. The museum says the degradation was reinstated after the war. Another chapter of the National Academies report describes it as set “at a very low level” during the war and during the initial occupation of Haiti.

A 1996 presidential directive then committed the US to stopping Selective Availability by 2006, with a yearly review, according to Clinton’s statement. The 2000 decision came six years early. On national security, the statement points to two findings: threat assessments concluded that setting the error to zero “would have minimal impact on national security”, and the US had “demonstrated the capability to selectively deny GPS signals on a regional basis”.

At the briefing, Money described that capability as the precondition: “we wouldn’t do so until we have an ability to deny it.” The military is also still testing ways to work without GPS at all. In October, the Navy described a drone that landed on a moving ship with no GPS signal.

How close it came to ten times

“Ten times” was a promise made on 1 May, before any data from the switch-off existed, and the fact sheet also describes the gain as an increase “by an order of magnitude”. The government’s own first-week measurement falls short of that. The Erlanger figures, 45.0 metres before and 6.3 metres after, work out to roughly a sevenfold improvement. That ratio is arithmetic on the government’s two numbers, not a figure the government published.

The Erlanger station also used a dual-frequency receiver at a fixed reference station, with a correction for the ionosphere, so its numbers are not what a dashboard unit showed in 2000. The 100-metre target in the National Academies report, for its part, was a ceiling written into a specification rather than a measurement of any receiver.

GPS.gov’s FAQ, last updated in October 2001, settled on a more modest claim: users around the world “should now be experiencing the same basic GPS accuracy of 10-20 meters or better”. White House science adviser Neal Lane put the change in plainer terms: “Before, you could be somewhat certain where something was with a couple of hundred feet. Now, you can pinpoint it precisely down to tens of feet.”

In September 2007, the White House announced that the next generation of satellites would be bought without the feature, so that Selective Availability “will no longer be present in GPS III satellites”. The government’s accuracy page says GPS-enabled smartphones are typically accurate to within 4.9 metres under open sky, citing a 2015 study.

GPS.gov says the US “has no intent to ever use SA again”, and its 2001 FAQ named what would take its place, saying the military is “dedicated to the development and deployment of regional denial capabilities in lieu of global degradation.”

Featured image: Artist’s impression of a GPS Block IIR satellite in orbit (representative; not a specific satellite). Illustration: US National Executive Committee for Space-Based Positioning, Navigation and Timing / Wikimedia Commons, public domain; resized.