The Ionosphere: Earth's Electric Mirror

What the ionosphere actually is

Climb high enough and the air stops being ordinary gas. Above about 60 km the atmosphere is so thin, and the sunlight so raw, that high-energy photons routinely knock electrons off atoms and molecules. The result is a partially ionized plasma: a soup of free electrons, positive ions, and — crucially — a majority of still-neutral atoms. This overlapping electrified region is the ionosphere, and it stretches from roughly 60 km up to around 1,000 km, where it grades into the plasmasphere and open space.

The most important word above is partially. Even at its densest, only a tiny fraction of the atmosphere is ionized. At the peak of the daytime F2 layer, electron densities reach about 10¹² electrons per cubic meter — which sounds enormous until you note that the neutral gas at that height is denser still, so less than 1% of atoms carry a charge. Down in the D region the ionized fraction is a few parts per million. The ionosphere is not a wall of plasma; it is a faint electric haze embedded in the thermosphere.

That distinction matters because two different "atmospheres" occupy the same volume. The thermosphere describes the neutral gas (its temperature climbs past 1,000 K, even ~1,500–2,000 K in solar maximum, because there is too little gas to shed the absorbed energy). The ionosphere describes the charged component living inside it. They interact constantly — neutrals collide with ions, winds push plasma around — but they are studied as distinct systems.

• Neutral thermosphere: what drags satellites and heats up in storms.
• Ionospheric plasma: what reflects radio, carries currents, and lights the aurora.

How sunlight builds and breaks it every day

The ionosphere is a balance sheet. On the production side, solar extreme-ultraviolet (EUV) and X-ray photons ionize atoms — mostly atomic oxygen, molecular nitrogen and oxygen, and nitric oxide. On the loss side, electrons and ions recombine, and negative ions form, wiping the ledger clean. The height where each layer peaks is set by a beautiful trade-off first described by Sydney Chapman in 1931: high up there is plenty of ionizing radiation but almost no atoms to ionize; low down there are atoms galore but the radiation has already been absorbed. The Chapman layer peaks where those two curves cross.

Because production tracks the Sun, the ionosphere breathes with the day. At sunrise the D and E regions switch on within minutes; at sunset they collapse. The D region — the great HF absorber — essentially vanishes at night, which is precisely why distant AM broadcast stations 'come in' after dark: the absorbing layer that muffled them by day disappears, and their signals can skip off the higher F region instead.

The F2 layer is the stubborn survivor. Its ions are atomic oxygen (O⁺), and at those altitudes the gas is so rarefied that recombination is slow — an O⁺ ion may wait hours to find an electron. So the F2 layer persists all night, thinner but never gone, keeping long-distance radio alive around the clock. It also does something counterintuitive called the Appleton anomaly: instead of peaking directly under the noon Sun at the equator, F2 density forms two crests roughly 10–20° north and south of the magnetic equator, because an eastward electric field lifts plasma up and it then diffuses down the magnetic field lines to either side — the 'fountain effect.'

Over the 11-year solar cycle, everything scales up and down. At solar maximum, stronger EUV output can raise F2 electron densities several-fold and push the peak higher, making higher radio frequencies usable; at solar minimum the whole structure sags.

The electric mirror: how it reflects radio

Here is the trick that made the ionosphere famous. A radio wave entering a plasma feels the free electrons, and the plasma has a natural plasma frequency set by electron density. A wave whose frequency is below the local plasma frequency cannot propagate through — it is refracted back, bent gently until it returns to Earth. A wave above it punches straight through to space. The dividing line, looking straight up, is the critical frequency (often written foF2 for the F2 layer), typically a few MHz at night to ~10–15 MHz by day at solar max.

Radio operators don't usually transmit straight up, though. Send a wave at a shallow angle and you can reflect frequencies well above the vertical critical value — the Maximum Usable Frequency (MUF) for a given path can be roughly 3× the critical frequency. This is skywave propagation: an HF signal in the ~3–30 MHz band bounces off the ionosphere, comes down thousands of kilometers away, can bounce again off the ground, and 'hop' its way around the planet. A single hop off the F2 layer can span 2,000–4,000 km; multi-hop paths circle the globe. Long before communication satellites, this is how ships, aircraft, broadcasters, and militaries talked across oceans.

• Below the MUF: the signal reflects and completes the hop.
• Above the MUF: the signal leaks into space (which is exactly what you want for satellites and radio astronomy).
• The 'skip zone': a dead ring between the ground-wave range and where the first skywave hop lands — you can hear a station 3,000 km away but not one 500 km away.

The mirror is also frustratingly fickle. A solar flare floods the D region with X-rays and causes a Sudden Ionospheric Disturbance — the Moegel-Dellinger effect, or 'Dellinger fadeout' — where HF communication on the sunlit side of Earth can black out for minutes to hours. And sporadic-E, patchy dense clouds in the E region, can suddenly reflect VHF signals that normally sail into space, letting FM and TV signals appear from impossibly far away.

Currents, storms, and the aurora

Because it is conducting, the ionosphere carries electric currents, and those currents connect it to the wider machinery of near-Earth space. During the day, tidal winds drag the plasma across Earth's magnetic field and drive the Sq (solar-quiet) current system, whose signature is a small daily wiggle in ground magnetometers. At the geomagnetic equator, a narrow ribbon of intense current called the equatorial electrojet flows eastward in the E region.

The real fireworks come during geomagnetic storms. When a coronal mass ejection or fast solar wind stream hits Earth's magnetosphere, energetic particles pour down the magnetic field lines into the upper atmosphere near the poles. There they collide with oxygen and nitrogen and make them glow — the aurora borealis and australis. Green light comes from atomic oxygen around 100–150 km (the 557.7 nm line); red from oxygen at higher altitudes near 200–300 km; blues and purples from ionized molecular nitrogen. The aurora is, quite literally, the ionosphere lighting up as it is bombarded.

Those same disturbances drive the auroral electrojets, currents that can carry on the order of a million amperes. By Faraday's law, a rapidly changing current overhead induces electric fields in the ground and in long conductors — power lines, pipelines, railways. This is how space weather reaches the ground: the historic 1859 Carrington event set telegraph offices sparking; the 13 March 1989 storm collapsed the Hydro-Québec power grid in about 90 seconds, leaving ~6 million people without electricity. Storm-time heating also puffs up the thermosphere, increasing drag and prematurely dropping satellites — a swarm of newly launched Starlink satellites was lost this way in February 2022.

Why it matters for GPS, and how we measure it

The ionosphere's mirror trick has a modern nemesis: satellite navigation. GPS, Galileo, and their kin transmit at ~1.2–1.6 GHz — far above any critical frequency, so the signals pass through easily. But 'easily' is not 'unaffected.' As a radio wave crosses the plasma it is slowed and bent in proportion to the number of electrons along its path, quantified by the Total Electron Content (TEC) — electrons in a column of one square meter cross-section, measured in TEC units (1 TECU = 10¹⁶ electrons/m²).

Typical vertical TEC runs from a few TECU at night to ~50–100 TECU by day, and each TECU adds about 0.16 m of range error at the GPS L1 frequency. Left uncorrected, the ionosphere is the single largest error source in single-frequency GPS, capable of tens of meters of position error during disturbed conditions. This is why precise receivers use two frequencies: because the delay scales as 1/frequency², comparing L1 and L2 lets the receiver solve for and cancel most of the ionospheric delay. Small, fast irregularities cause scintillation — rapid fading that can make a receiver lose lock entirely, a serious hazard for aviation near the magnetic equator and in the auroral zone.

We map all this with a purpose-built toolkit:

• Ionosondes — ground radars that sweep frequency and time the echoes to plot electron density vs. height (an 'ionogram'), the technique that discovered the layers in the 1920s.
• Incoherent scatter radars (Arecibo before its 2020 collapse; EISCAT; Millstone Hill) — powerful radars that read density, temperature, and plasma motion from faint scattered signals.
• GNSS receiver networks — thousands of GPS stations turned into a global, real-time TEC map.
• Satellites — from the topside sounder Alouette 1 (1962) to NASA's ICON (2019) and the COSMIC/COSMIC-2 radio-occultation constellations that profile the ionosphere from orbit.

A century of discovery

The ionosphere was predicted before it was seen. When Guglielmo Marconi bridged the Atlantic by radio in December 1901, physicists were baffled: radio waves travel in straight lines, so how did a signal follow the curve of the Earth? In 1902, working independently, Oliver Heaviside in England and Arthur Kennelly in the United States proposed that a conducting layer high in the atmosphere was bouncing the waves back down. For two decades this Kennelly–Heaviside layer remained a plausible but unproven hypothesis.

Proof came in 1924–1925. In Britain, Edward Appleton and Miles Barnett varied a transmitter's frequency and watched the received signal beat between the direct ground wave and a delayed reflected wave, measuring the reflecting layer's height at about 100 km and later finding a higher layer near 300 km — the layer now named after him. Almost simultaneously in the U.S., Gregory Breit and Merle Tuve did it with sharp radio pulses, timing the echo directly — the direct ancestor of both the ionosonde and radar. Appleton received the 1947 Nobel Prize in Physics for this work; he also coined the lettered layer scheme (D, E, F) that we still use, having simply started at E for the electric field of the reflecting layer and left room below and above.

The name 'ionosphere' was suggested by the Scottish physicist Robert Watson-Watt around 1926 — the same man who would, within a decade, turn radio reflection into radar. From those pulse experiments grew an entire science, and a global research campaign: the International Geophysical Year (1957–58) put ionospheric physics at center stage, coinciding with the launch of Sputnik 1, whose radio beeps were themselves tracked partly through the very layer this story began with. A hundred years on, we still watch Earth's electric mirror every day — now to protect the satellites and power grids that a Marconi never imagined.

The main ionospheric layers, their altitudes, ionizing sources, and radio behavior. Densities and heights vary strongly with time of day, season, and solar activity.

Layer	Approx. altitude	Ionized by / behavior	Radio effect
D region	60–90 km	Solar Lyman-α & hard X-rays; ionizes NO. Vanishes at night.	Absorbs HF (daytime AM fadeout); Lyman-α on NO
E region	90–150 km	Soft X-rays & EUV; molecular O₂⁺, NO⁺ ions. Weak at night.	Reflects lower HF; sporadic-E gives surprise long hops
F1 region	150–220 km	EUV; present mainly on the dayside, merges into F2 at night	Minor reflector; often blends with F2
F2 region	220–400 km	Densest layer; atomic O⁺ ions; persists all night	Main long-distance HF mirror; sets the MUF