Module 4 of 6 · 7 min

How forecasting works

Three instruments, in order of warning time: coronagraphs that see the launch days out, a physics model that turns launch speed into an arrival, and the L1 monitor that turns the last hour from a forecast into a measurement.

One CMEThe main Gannon CME, mid-transit. Selected wedge, arrival drawn at Earth.

The bright wedge is the selected CME, a few hours after launch. Its arrival estimate moves when you change the wind below.

The idea

A fast CME plows into the slower solar wind ahead of it and decelerates, like a boat pushing through water. Feed the model a launch speed, a direction and the wind it must push through, and physics gives an arrival time. The catch is that the wind changes daily and a head-on CME's speed is hard to measure. That is where the error bars come from.

The launch speed and direction come from coronagraphs: telescopes with a disk that blots out the Sun, so the faint cloud around it shows. What they see depends on where the cloud is going.

Wide coronagraph view of the August 31, 2012 CME as a large pale loop spreading to the left of the Sun — Real imageGoing sideways. A cloud heading off to one side shows as a loop, and its speed can be read straight off the pictures. This one, from 2012, was easy to measure. Images: SOHO LASCO (ESA and NASA), NASA SDO.

Difference image from the SOHO coronagraph on May 9, 2024: a pale ring of new material surrounding the hidden Sun on every side — Real imageComing at us. The CME on the map above, on May 9, 2024, seen from near Earth. A cloud heading for the camera shows on every side of the Sun at once: a *halo*. It is the dangerous kind and the hardest to clock, because it is coming straight down the line of sight. Shown as the change between two frames 48 minutes apart, which is how forecasters bring a faint halo out. Image: SOHO LASCO C3 (ESA and NASA), via Helioviewer.

Turn the dial

Same CME, same launch. Change only the wind it must push through and watch the arrival move.

Our modelThe same CME as on the map, run through the wind you picked. The dashed line is where it would be with nothing in its way. Distances are to scale.

Loading the forecast…

Aim has two parts

One more thing decides a forecast before any arrival time matters: does it hit at all? Here is another CME from the same week, alone, seen from straight above the solar system, the way the flat map shows it. Your call: will it hit Earth?

3DOne CME, launched May 10, 2024, seen from above.

Drag to look around. The faint shape on the grid is the cloud's footprint on the plane of the planets.

Check yourself · Module 4

1. The L1 monitoring spacecraft give roughly how much certain warning?

L1 sits about 1.5 million km upstream. When a shock sweeps past it, arrival at Earth is measured, not modeled: half an hour to an hour out.

2. Why is a storm's strength so much harder to predict than its arrival time?

Arrival is physics on speed and distance. Strength hinges on whether the CME's field points south, invisible until the last hour.

3. Our published arrival accuracy is honestly about…

Twelve hours at the median, about fifteen on average, validated against hundreds of real arrivals. Imperfect inputs, honest physics.

← Module 3: Watch a storm happen Module 5: Predict it yourself →

How forecasting works

The idea

Turn the dial

How often is it right?

Aim has two parts

Check yourself · Module 4