Chapter 1, Section 4

Why the Bottom of the AM Band Dies First

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

Two stations, same power, same market: the one lower on the dial is far harder to keep alive. The reason is buried in the ground, literally, and almost nobody outside a station engineer ever explains it.

When an AM station goes dark, the obituary usually blames the economics, and the economics are real. But there is a quieter factor that decides which stations die first, and it has nothing to do with programming or ratings. It is the frequency itself. A license at the bottom of the AM band is far harder and more expensive to keep on the air than one at the top, and when the culling comes, the low end thins out first. Here is the physics behind it.

It starts with wavelength. An efficient AM antenna wants to be about a quarter-wavelength tall, and wavelength runs inversely with frequency: the lower the frequency, the longer the wave, the taller the tower. At 560 kHz a quarter-wave is roughly 440 feet of steel; at the top of the band, 1600 kHz, it is only about 150 feet. That alone makes a low-band station a bigger, costlier, more zoning-fraught proposition. But the tower is not the part that kills these stations. The land is.

The real killer is the ground system, not the tower. Every AM tower sits at the center of a buried copper “ground screen”: the FCC standard is 120 radial wires, each about a quarter-wavelength long, fanning out in a circle. Here is the trap: the tower height scales linearly with wavelength, but the radial field is a disc, so its area scales with the square of the wavelength. Drop from 1120 kHz to 560 (exactly one octave) and the tower merely doubles in height, but the buried copper field quadruples in area.

Frequency¼-wave towerRadial lengthGround field, per tower
560 kHz~440 ft~440 ft~14 acres
630 kHz~390 ft~390 ft~11 acres
720 kHz~340 ft~340 ft~8.4 acres
1120 kHz~220 ft~220 ft~3.5 acres
1600 kHz~150 ft~150 ft~1.7 acres

Table 1.4-1

Side view: the tower~440 ft560 kHz~220 ft1120 kHz~150 ft1600 kHzTop view: the buried copper radials around one tower~14 acresradials ~440 ft long~3.5 acresradials ~220 ft long~1.7 acresradials ~150 ft long200 ftone scale forthe whole drawing560 kHz is one octave below 1120 kHz: twice the tower height, four times the landOnly 30 of the 120 radials are drawn. Figures are for one non-directional tower.

Drawing 1.4-3

Those figures are per tower, for a single non-directional stick. Most big-market low-band stations run a directional array of several towers, each with its own radial field, and each spaced from the next by a fraction of a wavelength that is also larger at the low end. Stack the overlapping radial screens and the spacing, and a station like the old KSFO 560 sprawled across an eleven-acre site. That is the real estate a low-band AM demands: not a rooftop, but a small farm’s worth of unobstructed, high-conductivity ground: ideally coastal salt marsh, because low-band ground-wave coverage lives or dies on soil conductivity.

Why it can barely move. When a high-band station loses its site, it has an escape hatch: diplexing: sharing another station’s existing tower through a filter network. A 1500-kHz orphan can find a host almost anywhere, because even an ordinary 200-foot tower is electrically tall enough at high frequencies. A 560 or 630 orphan needs a host tower that is physically enormous, and those are rare and almost always already occupied. Worse, if the low-band station runs a directional pattern to protect its co-channel neighbors, it cannot just borrow one tower: it needs a whole host array with the right geometry, which effectively does not exist on the open market. High up the band, relocation is a phone call. Down at the bottom, it is a dead end.

Three deaths, one disease. Watch it play out on the West Coast dial in the space of a single decade: three stations, three cities, three different business stories, one physical truth underneath.

StationFreqHow it died
KIDD
Monterey
630Fifty-year-old towers rusted out on land that had become protected coastal wetland; new towers could not be erected there, and a borrowed single tower could not reproduce the directional pattern the rules required. Silent 2014; license surrendered 2023.
KDWN
Las Vegas
720The owner sold the tower-site land to a warehouse developer for $40 million, then tried to relocate onto sister station KXNT’s site, and could not finish before the one-year silence deadline. Silent 2023; license surrendered 2024.
KZAC
(ex-KSFO), San Francisco
560A century-old 5-kW signal on leased Islais Creek marsh. Once the money-making format moved up to 810, the owner moved to reject the tower lease rather than sell. Silent 2025; license lapsed 2026.

Table 1.4-2

Each sat on a large, specialized, low-band site that could not be economically replaced, and each proved the same thing: once the signal leaves that ground, a bottom-of-the-band license has nowhere to go. The KSFO 560 story is told in full on its own page; KIDD and KDWN are the same story with the serial numbers filed off.

The one-year guillotine. Federal law then makes it final. Under Section 312(g) of the Communications Act, a broadcast license expires automatically the moment a station passes twelve consecutive months off the air: no hearing, no order, no ceremony. For a low-band station that cannot re-site, that clock is a guillotine: the day the transmitter goes quiet, a one-year fuse starts burning, and there is rarely a way to beat it. And the FM translator many of these stations now feed makes surrender painless: the audience keeps the programming on FM while the expensive AM towers come down.

The regulators have noticed. This is no longer hypothetical. The National Association of Broadcasters is now pressing the FCC to eliminate the minimum antenna-efficiency standards that, in the NAB’s own words, keep AM stations from “locating antennas on smaller, less expensive parcels that are closer to their audiences.” Read that plainly: it is a push to let AM stations use less land. The minimum-efficiency rule is exactly what forces the tall tower and the sprawling radial field; relax it, and a station can put a smaller, less-efficient antenna on a modest lot. But there is a devil’s bargain inside: less land is bought with less signal. A 560 on a small parcel with a stubby antenna barely covers its own city; yet a barely-covering station that survives beats a silent one whose license has already lapsed.

The bottom line. The economics decide whether anyone wants to save a station. The physics of the low band decide whether they can. That is why, as the AM band slowly contracts, the bottom of the dial goes first, not because those signals matter least, but because they are the ones nobody can afford to keep, rebuild, or move. The great exception is the low-band clear-channel giant, a KFI 640 or a WSM 650, that owns its land outright and whose 50,000-watt reach is still a genuine asset; those may outlast everything around them. But the ordinary regional station at the bottom of the band is running out of ground, in every sense.

Sources: FCC AM technical rules, 47 CFR §§ 73.45 & 73.189 (antenna systems / minimum efficiency) & 47 U.S.C. § 312(g) · RadioInsight (KIDD Monterey Goes Dark; Audacy Surrenders KDWN & KXST) · Bay Area Radio Museum (KSFO / Islais Creek) · Radio World & Inside Radio (NAB AM revitalization / antenna-efficiency filing) · N6JET.

More AM Radio Stories

AM Radio Facts and Trivia, the main page, with hundreds more records, oddities and firsts. Test yourself with the AM Trivia Challenge.

More in The Physics of AM

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

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