Sharper GPS needs even more accurate atomic clocks

The GPS network might just be Earth’s greatest piece of infrastructure. It’s effectively a collection of clocks in space that serve up time information 24/7 free of charge to anyone on the planet who cares to listen.

These timing signals have all sorts of important applications, but most people will be using them to help get from A to B with the aid of GPS navigation tools. Well, actually to within about 10 metres of B. On the scale of humans and cars that’s a fairly substantial margin of error, as anyone who’s missed that left turn can attest to.

Up and coming technologies such as self-driving cars will probably rely upon a combination of local sensing and GPS signals to navigate independently without incident. So any improvement in GPS accuracy would be hugely advantageous in speeding up and rolling out the era of the autonomous car.

A GPS receiver listens and compares the different timing signals from the GPS satellites and then uses that information to calculate exactly where on Earth you are.

A variety of factors conspire against the accuracy of GPS navigation. Right now, for most civilians, the primary offender is the Earth’s ionosphere, which interferes with the timing signals as they commute from a satellite to your GPS receiver of choice.

But the second biggest contribution of error comes from the stability of the clocks onboard the GPS satellites.

Timing is everything

Every single GPS satellite is home to a family of atomic clocks (typically four) that derive their time from cesium (Cs) or rubidium (Rb) atoms.

What is actually being measured in these clocks is the energy difference between two specific atomic states. When an atom changes from the high-energy state to the lower energy state, the energy difference is emitted in the form of light. The frequency, or ticking rate, of this light is what we count and how we define time. The crucial part is that, fundamentally, this energy difference is always the same.

All clocks – be they wrist, grandfather, atomic or otherwise – have some level of intrinsic error that causes them to lose seconds or drift away. Left to their own devices, the atomic clocks on board the GPS satellites would drift, meander or dawdle about by 10 nanoseconds a day.

That may not sound like much, but if you go ahead and multiply by the speed of light, you arrive at a GPS position error of three metres.

Thankfully, the GPS network is well monitored and corrections are applied to keep the clocks in-line, so that they are only responsible for about one or two metres of position error.

There is lots of work underway to improve the accuracy of GPS navigation, including different broadcast methods that can effectively eliminate the influence of the ionosphere.

But, ultimately, the performance of the clocks is fundamental to GPS. Clock technology is advancing all the time and with it comes lots of new opportunities for discovery and applications.

Improved accuracy

Here at the University of Western Australia our research group is building an optical lattice clock based on ytterbium (Yb) atoms.

In addition to the ytterbium atoms, we also have an ultra-stable laser and a frequency comb: all the necessary components to produce an incredibly accurate optical atomic clock.

The UWA ytterbium lattice clock. Romain Bara-Maillet and John McFerran, Author provided

This type of clock is currently one of the best that you can make, with similar designs elsewhere achieving accuracies 100,000 times better than what you would find on a GPS satellite.

They are purported to be accurate at the level of 10^-18 of a second. If two such clocks started running when the universe began 13.8 billion years ago, to this day they would agree to within 1 second.

When complete, the lattice clock will be the only one of its kind in the southern hemisphere. The clock, along with our other cutting-edge time and frequency technologies, will form a ground-station at UWA for participating in space-clock comparison experiments.

In fact, we need this impressive clock in Australia to play a crucial role in an upcoming European Space Agency experiment.

Missions in space

The Atomic Clock Ensemble in Space (ACES) mission will place a different type of cold-atom clock on board the International Space Station (ISS), one that is about 1,000 times more stable than your typical run-of-the-mill GPS atomic clock.

The ACES mission is on track for a 2017 launch. Successfully getting this thing up in to space and operating on the ISS without incident will be a pretty significant achievement by itself. It is an important stepping stone on the path towards setting up future space-clock networks.

Over the course of a few years the timing signal produced by the ACES clock will be compared against different types of clocks all over the world, including the ytterbium clock under development at UWA.

This will allow us to undertake some important tests of fundamental physics, such as testing gravitational redshift and searching for subtle changes in the fundamental constants of nature.

Outside of space missions there are still a whole bunch of things you can do with an extremely stable and accurate optical lattice clock. For example, as the ticking rate of these clocks is strongly dependent upon the strength of the local gravitational field, two clocks separated by 30cm height already run at noticeably different rates.

This could eventually lead to a network of such clocks being used to accurately map out the Earth’s gravitational field, which could be useful for minerals exploration.

Even though scientific progress can sometimes feel a bit slow, it’s only a matter of time before advanced clocks are going to be incorporated into upgraded GPS satellites and helping to accurately drive your (presumably autonomous) car here, there and everywhere.

Work discussed in this article is partially funded by the Australian Research Council.