The Magnetometer in Your Pocket: Turning a Smartphone Into a Space Weather Station

Your phone already knows which way is north. What is less widely appreciated is that it knows this by measuring the strength and direction of a magnetic field, which means it is — in a limited, pocket-sized way — a scientific instrument. And the thing it is measuring is not perfectly steady. Solar wind buffets it, storms distort it, and on a good night the whole field over Britain twitches in ways a careful amateur can actually see.

What is actually inside the handset

Behind the screen sits a magnetometer: a chip a few millimetres across, usually a Hall-effect or magneto-resistive sensor, with three perpendicular sensing axes so it can report a field vector rather than a single number. Typical parts can resolve changes of a fraction of a microtesla and measure fields up to a few thousand microtesla, comfortably covering everything from Earth's field to the magnet on your fridge door.

It is there for mundane reasons. A compass heading makes map apps usable, helps augmented reality know which way the camera is facing, and lets the phone work out its orientation when GPS is struggling. Nobody fitted it for science. That does not stop it being useful.

What it can and cannot feel

Earth's magnetic field over the UK is around 48 to 50 microtesla in total strength, but most of that points steeply downwards. The horizontal component — the part that actually steers a compass needle, and the part that space weather disturbs — is only about 20 microtesla. A serious geomagnetic storm might shift that by a few hundred nanotesla, which is to say a few tenths of a microtesla.

That is right at the edge of what a phone can resolve. The sensor's raw noise, the metal in your desk, the speaker in your handset and the magnets in a phone case all produce offsets far larger than a storm signal. So no, your phone will not replace an observatory. But with patience, a bit of averaging and a quiet spot, it can pick out the big events — and the process of trying is genuinely good science.

Some practical wrinkles first. Smartphone compass apps usually show a crafted heading, not raw field data, so you need an app that exposes the sensor. Android handsets generally give developers access to calibrated and uncalibrated magnetic field readings; iPhones are more restrictive but apps such as phyphox can still log values. Both platforms benefit from a figure-of-eight wave of the hand now and then, which helps the phone recalibrate.

Three experiments for a wet afternoon

1. Survey your kitchen

Open a sensor-logging app and walk slowly around the room, watching the total field strength. Fridge doors, oven fronts, knife strips and the little magnets holding a shopping list will all announce themselves. This is a superbly cheap way to learn what a magnetometer actually responds to — and why a phone lying next to a laptop charger is useless for measuring the planet.

2. Plot a magnet's field against distance

Tape a small magnet to a ruler and log the field strength at 2, 4, 6, 8 and 10 centimetres. Graph it. If you plot the logarithm of field strength against the logarithm of distance, the gradient should come out close to minus three for a simple dipole. It is a real measurement of a real physical law, done with hardware you already own.

3. The overnight storm watch

This is the one worth setting up properly. Put the phone in airplane mode, leave it on a wooden table at least half a metre from anything metal or electrical, and start a logging app that records the magnetic field at one sample every few seconds. Run it from dusk until morning. Export the data as a CSV and plot it.

Then comes the interesting part: compare your trace with a professional magnetogram. The British Geological Survey publishes near-real-time traces from its observatories at Eskdalemuir, Hartland and Lerwick, and those give you a clean, calibrated record of the same event. If a geomagnetic disturbance shows up in Scotland and your kitchen table shows a matching wobble at the same minute, you have done citizen science.

Reading the sky like an observatory

A few sources are worth bookmarking. The BGS space weather pages carry magnetograms and storm commentary. The Met Office's Space Weather Operations Centre issues forecasts and alerts from Exeter. AuroraWatch UK, run from Lancaster University, sends aurora alerts based on magnetometer data — useful even when cloud cover over Britain makes the actual lights a distant hope.

The Kp index is the headline number most people quote, but it is a planetary average; the local magnetogram is what tells you whether the field above your postcode actually moved. Big storms, like the one that painted skies across the UK in May 2024, show up as sharp, unmistakable deviations. Quieter disturbances are subtler and demand better instrumentation than a phone.

What the professionals do differently

A proper magnetic observatory looks nothing like a kitchen table. Sensors sit in non-magnetic huts, sometimes buried, sited well away from roads, railways and power lines. The instruments are fluxgate magnetometers for vector data, often paired with proton precession magnetometers for absolute total-field values, and a DI fluxgate theodolite for occasional absolute determinations of declination. Data is sampled by the second and calibrated against reference standards, so that a measurement taken in 1997 is directly comparable with one taken today.

That absolute stability is the thing a phone fundamentally cannot provide. Your handset's sensor drifts with temperature, carries factory offsets, and is nudged by whatever the operating system feels like doing. It is a detector of change, not a keeper of standards.

Be honest about the error bars

A few habits will save you from embarrassing conclusions. Always log for at least an hour before trusting a baseline, because phone sensors settle. Note the room temperature if you can — thermal drift is real. Never compare a raw phone reading with a published absolute value and expect agreement; compare patterns in time instead. And if it looks like you have detected a storm, check whether the washing machine was on.

None of that diminishes the appeal. Somewhere in the BGS archives is a century of meticulous recording, and somewhere in your pocket is a sensor small enough to ignore, watching the same field. The gap between them is mostly calibration and patience — and both are learnable.