The surface where the solar wind’s pressure balances Earth’s magnetic field. Its shape follows Shue et al. 1998, driven by two measured quantities: the dynamic pressure of the wind and the north–south component of the interplanetary field. Both come from NOAA’s propagated solar wind, so the surface you see is the one this moment’s wind is making.

The nose sits around ten Earth radii. A strong gust can push it inside geosynchronous orbit at 6.6 Re, which is why that distance matters.

Where the solar wind first slows below its own signal speed and piles up. It is not a second surface with its own physics: it is the magnetopause scaled outward by the magnetosonic Mach number, so it shares the shape and adds only a second nose distance.

That Mach number has its own inputs — speed, density, temperature and field strength — and if any of them is missing there is no shock at all. It does not fall back to a guess: the formula has a numerical floor at Mach 1.2 that a missing measurement would slip through as though it were real, putting the shock nearly four times too far out without a single warning.

Between it and the magnetopause lies the magnetosheath, the slowed and deflected wind — which is what the solar wind particles are flowing through when they bend around the nose.

Traced through the field itself: IGRF-14 for the Earth’s own magnet, T89c for what the currents in space add to it, driven by the measured Kp of this moment. Each line starts just above the surface and is followed until it either comes back down or crosses the magnetopause.

That crossing is the whole point, and it is what the three colours mean. A closed line has both feet on Earth. An open line reaches the boundary drawn around it, so plasma can travel along it either way. Unresolved is neither — the integration ran out of room or out of steps, and that is a statement about the tracer and not about the magnetosphere, which is why those lines are thinner and dashed. The full key is in Details.

Lighter and warmer lines leave the northern hemisphere, darker and cooler ones the southern. That is a shade within a state, not a fourth state.

A grid in Earth radii, one line every 10 Re, in the plane the view looks at. It replaces the star field in Meridian and Top: those are cross-sections, not a view from somewhere, and a sky behind them would say otherwise. The free 3D view keeps its stars and has no grid — a scale needs a flat plane to mean anything.

Where the field would lie if nothing were pushing on it — a pure dipole, drawn from the textbook identity r = L·cos²(latitude) about this moment’s magnetic axis. No solar wind, no Shue, no T89, and no fitted constants: it is the zero-wind limit rather than an average of quiet days.

It uses the same shells as the traced lines, so the two families share their feet on the surface and part company from there. That parting is the whole point: it is the measure of how far this moment’s field has been pushed out of shape. The outermost shells would reach 23 and 92 Earth radii undisturbed, which is why they run off the edge of the drawing.

Always drawn as a flat slice through the noon–midnight plane, even in the 3D view, and dashed — what comes out of a formula should not look like what was traced through a measured field.

The wind arriving from the Sun, drawn as it flows around the cavity. Two things about it are measured: how fast it moves and how many particles there are. The speed on screen scales with the measured velocity, so 700 km/s really does travel 1.75× as fast as 400, and the density sets how thick the stream is.

The path is drawn, not solved. Outside the bow shock each particle follows the measured inflow vector; inside it, the flow is deflected around the magnetopause by a formula, not by magnetohydrodynamics. What is true in it is the direction: the stream really does bend around the nose rather than run into it, and it brightens where it grazes the boundary, because that is where the pressure is highest.

No standoff, no wind — there would be nothing to bend around, and a straight stream through the magnetosphere would claim the opposite of what happens.

Two weather satellites on the geostationary belt, 6.62 Earth radii out. They are the only place in this scene where an instrument actually measures the field — everything else you see is IGRF, T89 or Shue saying where the field ought to be.

The number beside each one is the external field: what the magnetometer read, minus the internal field the model puts at that spot. That subtraction is exact, so what is left is ring current, tail current and magnetopause currents — the part that actually varies. It is the same figure as the GOES external row in Details, read from the same sample.

The marker is usually hollow, and the label says by how much. A filled dot would claim the craft sits exactly where you see it, and in a cross-section it almost never does: the belt lies on Earth’s equator, while both fixed views cut through planes defined by the magnetic axis and the Sun line.

The two views differ in how far off it is, not in whether. Over a full day, a craft sits a median 4.7 Re out of the Meridian plane and 1.2 Re out of the Top plane — so Top is about four times closer to the truth, and the number beside the marker is what tells you that.

When a label says 4 min old, that is the age of the sample rather than a delay in the app. NOAA publishes the magnetometers a few minutes behind the wall clock, while the solar wind feed is propagated to arrival time, so at “now” the newest GOES sample is typically two to six minutes old. It is used anyway, up to ten minutes: ring current and tail current move over hours, so a sample a few minutes old is the same sample. Past ten minutes the number drops out rather than going stale quietly. Below two minutes no age is shown, because there is nothing to warn about.

A marker marked arcjet changes colour and gains a ring: the satellite’s own thrusters were firing, and the magnetometer was partly measuring them. Those samples are flagged rather than dropped — leaving one out would be a claim of its own.