Chapter 1, Section 2

Groundwave, Skywave & the M3 Map: Why the Dial Changes at Night

Part of AM Radio Facts and Trivia: The AM Broadcast Band in the USA. Updated November 15, 2026.

Almost every number on this site (the power drops at sunset, the towers that multiply after dark, the protected clear channels, KFBK's record field strength) descends from one set of facts about how medium-wave radio actually travels. It's worth stating them plainly in one place.

Groundwave: the daytime workhorse. By day, an AM signal clings to the curve of the earth and propagates as a groundwave. How far it reaches depends overwhelmingly on what it's traveling over, because the ground itself is part of the circuit. This is measured as ground conductivity, in millisiemens per meter (mS/m): poor, dry, rocky soil might be 1 mS/m; rich farmland or marsh 15-30; and seawater an astonishing ~5,000. That single variable is why a Gulf Coast or river-delta station throws a groundwave far out of proportion to its power, and why the note on the KFBK 1530 page reads "Ground Conductivity = 15", fifteen millisiemens of cooperative California valley soil under a record-setting Franklin antenna.

The M3 map. The FCC publishes the conductivity of the entire country as a contour map, the famous M3 map, and engineers lay their proposed coverage right on top of it. It is, in a real sense, the soil-quality atlas of American radio: the same 50,000 watts covers wildly different amounts of ground depending on where the M3 map says you're standing.

Skywave, and why night is different. Sunlight charges a low layer of the ionosphere called the D-layer, which by day absorbs medium-wave signals that try to bounce skyward. At sunset the D-layer dissipates, and the higher E and F layers begin reflecting AM signals back to earth hundreds or thousands of miles away. That night-time skywave is a gift to listeners (see Chasing the Skip) and a curse to the band: suddenly every station on a frequency can hear every other one. The entire apparatus of nighttime power cuts, directional arrays, clear-channel protection, and the vanished daytime-only stations exists for one reason: to manage skywave interference after dark. It's why a station can be a local lion at noon and drop to a single watt at midnight.

Sources: FCC 47 CFR 73.183-73.190 (groundwave/skywave; M3 conductivity map) · NAB Engineering Handbook · "The Development of the Directional AM Broadcast Antenna" (Radio Heritage Foundation).

WHY THE AM DIAL CHANGES AT NIGHT: GROUNDWAVE vs SKYWAVE DAY D layer groundwave: hugs the earth skywave absorbed: daytime D layer soaks it up daytime coverage = groundwave only; how far depends on ground conductivity (see the M3 story in the text) NIGHT F layer D layer: gone after sunset groundwave reflected! lands 500+ miles out skip zone: neither wave reaches here same transmitter, same power, suddenly heard two states away. This is why night rules exist.

Same station, two different physics. By day, an AM signal travels as groundwave: it hugs the curve of the earth, reaching as far as the ground conductivity allows, while the sun-charged D layer of the ionosphere acts as a sponge and soaks up anything radiated skyward. After sunset the D layer evaporates within minutes, uncovering the higher E and F layers, which act as a mirror: the skywave that died at noon now bounces and comes down hundreds of miles away at nearly full strength. Between where the groundwave fades and where the skywave lands lies the skip zone, close enough to "know" the station, too far for groundwave, too near for the bounce. This nightly transformation is the engine behind almost everything else on this site: clear channels, directional night patterns, daytimers, power cuts at sunset, and the DXer's whole hobby.

Drawing 1.2-1

Interactive · Groundwave

Ground Conductivity & Frequency Explorer

Two stations, same power. The one with the worse antenna can cover far more ground. Set the soil they're transmitting over and see why the bottom of the dial wins.

The ground beneath both stations

Station A

Station B

Drawing 1.2-2

Distance to contour5 mV/m
city grade
2 mV/m
primary
0.5 mV/m
fringe
Station A
Station B

Table 1.2-3

The model. Launch: sinusoidal-current monopole over perfect ground; radiation resistance by numerical integration, field at the horizon = 60·√(P/Rm)·(1−cos G). A quarter-wave comes out at 36.6 Ω and 313.8 mV/m per kW at 1 km; peak field lands at 228°, which is Ballantine's 5/8-wave optimum. Travel: Norton flat-earth groundwave attenuation over the numerical distance p. Cross-checked against the FCC license figures elsewhere on this page: WHO's 471.54 mV/m per kW scales to 3,334.3 mV/m at 50 kW, matching the licensed 3,334.29. It is a smooth-earth approximation: no terrain, no mixed paths, no skywave. Real coverage is messier. 73 de N6JET

Interactive · Propagation

Why the Dial Changes at Night

One station, one power, one tower. Scrub the clock. The D layer builds at dawn and swallows everything the tower throws upward; at sunset it evaporates and the same signal comes down two states away. Every power cut, every night pattern, every daytimer on this site exists because of what you're about to watch happen.

Sun elevation Sunrise Sunset D-layer absorption

Cross-section: the signal leaving town

Drawing 1.2-4

Elevation pattern: where the power goes

Drawing 1.2-5

Where the signal landsGroundwaveSkywave
Reaches the 0.5 mV/m contour at
Field at 300 miles
Takeoff angle for that one hopn/a

Table 1.2-6

What's exact and what isn't, because it matters. Exact: the elevation pattern, which is the same sinusoidal-current monopole integral used by the other explorers on this site; and the one-hop geometry: takeoff angle and slant path from spherical trigonometry with a 6,371 km earth. Sweep the tower past about 190° and watch a second lobe climb into the sky: that is the mechanism behind the WHO 1040 story, though WHO's actual sectionalized "shortened Franklin" is not a plain monopole and is not modelled here. Calibrated, not derived: the absolute skywave level and the D-layer absorption, which uses a Chapman-style (cos χ)0.75/f² scaling tuned so that a 50 kW station lands near the FCC's 0.5 mV/m clear-channel contour at continental distances. It is not the FCC skywave curves of 47 CFR 73.190, and it should not be used to plan a station. Solar geometry is local solar time, not clock time, and ignores refraction. Real skywave also fades, splits, and arrives by multiple hops, none of that is here. 73 de N6JET

© 2026 Chris, N6JET. Free to read and share for non-commercial use with credit (Creative Commons BY-NC 4.0). Not for sale.

More in The Physics of AM

All AM radio stories by subject · AM radio interactive tools and simulators · AM Trivia Challenge quiz