Why can you sometimes work Australia with 100 watts and a wire, while on other days the next country is hard to reach? The answer lies 60 to 500 km above your head. This guide explains how radio signals travel around the world, what the Sun has to do with it, and how HAMIOS turns live measurements into band advice.
Radio propagation has its own vocabulary. Here are the terms this guide uses, each in one sentence — come back here whenever a word is unclear.
The atmosphere tells two stories. The lower one — weather, temperature, the ozone layer — matters for VHF and UHF. The upper one, the ionosphere, is where shortwave (HF) propagation happens: a thin plasma of free electrons, created by the Sun, that bends radio waves back to Earth.
High above the weather, sunlight turns a thin layer of air into a kind of mirror for radio waves: the ionosphere. Shortwave signals bounce off it and come down hundreds or thousands of kilometres away. The weather layers below it only matter for VHF and UHF.
All weather happens in the troposphere, the lowest 10–15 km. Shortwave passes straight through it, but for VHF and UHF it matters a great deal: a temperature inversion — warm air on top of cold air, typical of high-pressure weather — can trap 2 m and 70 cm signals in a duct and carry them hundreds, sometimes over a thousand kilometres (“tropo”). The stratosphere above it, with the ozone layer, plays no role in radio propagation.
Above roughly 60 km, extreme ultraviolet (EUV) and X-rays from the Sun knock electrons loose from atoms and molecules. The free electrons form a plasma that refracts radio waves: the denser the plasma, the higher the frequency it can bend back. A layer with peak density N reflects frequencies up to f = 9·√N hertz when you transmit straight up — its critical frequency (foF2 for the F2 layer).
Because the ionisation comes from the Sun, the ionosphere changes all the time: with the time of day, the season, the 11-year solar cycle, and sudden events such as solar flares and geomagnetic storms. Reading those changes is what lets you find an open band instead of calling CQ into silence.
The ionosphere is not one uniform mirror but a stack of regions at different heights, each created by a different part of the solar spectrum and each with its own effect on your signal.
The mirror has floors. The lowest one (D) is a sponge: by day it soaks up signals on the low bands. E and F1 are small mirrors. F2, the highest, is the big mirror that makes long-distance (DX) contacts possible — day and night.
The D region forms at sunrise and fades within an hour or so after sunset. It is ionised mainly by the Sun's Lyman-alpha line (acting on nitric oxide) and by hard X-rays. The air there is still dense, so the free electrons keep colliding with molecules: instead of refracting a radio wave they absorb its energy. The D region does not reflect — it absorbs.
Absorption falls with the square of the frequency, so the low bands are hit hardest. By day, 160 m and 80 m barely reach beyond ground-wave and NVIS range; 40 m is limited to shorter distances; from 20 m upward the loss is small. After dark the D region disappears and the low bands open for DX. A strong solar flare can make it so dense that even 20–10 m fade out on the sunlit side of the Earth.
The E region peaks at about 105–120 km and is ionised by soft X-rays and EUV. It follows the Sun closely: strongest at local noon, weak at night. Its critical frequency foE is typically 3–4 MHz at midday — enough for single hops of up to about 2000 km on 40 m and sometimes 20 m.
Sporadic E (Es) is its famous exception: thin, very dense clouds of ionisation at 100–120 km that come and go within minutes to hours. They form when wind shear in the upper atmosphere squeezes long-lived metal ions — left behind by meteors — into a thin sheet. Es reflects 10 m, 6 m and sometimes even 2 m over 500–2300 km. It peaks from May to August in the Northern Hemisphere, with a smaller peak around midwinter. HAMIOS detects Es from the measured foEs at ionosondes and from short-skip spots on 10 m and 6 m.
On summer days — and more often in years of low solar activity — a ledge appears below the F2 peak: the F1 layer. It is ionised by EUV and follows the Sun like the E region. At night and in winter it merges into a single F layer.
For most operators the F1 layer is a side note: the F2 layer above it usually sets the MUF. Occasionally, on medium-distance paths, the F1 layer is the one that reflects.
The F2 layer is the one DX depends on. It is the densest layer, peaks at about 250–350 km and, because the air up there is so thin, electrons recombine slowly: the layer survives the night, weakened. Its critical frequency foF2 ranges from 2–4 MHz on a quiet night to 8–13 MHz on a sunny day near solar maximum.
One F2 hop covers up to about 4000 km (at very low take-off angles); longer paths use several hops. Every hop costs a few dB — the ground reflection most of all — so CW and digital modes such as FT8, with 10–25 dB more margin than SSB, reach much further with the same station.
It is also the most variable layer. It follows the solar cycle (higher sunspot number, higher foF2), the season (in winter, daytime foF2 at mid-latitudes is often higher than in summer — the winter anomaly) and geomagnetic storms, which can lower foF2 by 30–50 % for a day or more.
A signal can reach another station in several ways. Which one works depends on the frequency, the take-off angle, the time of day and the state of the ionosphere.
Your signal can reach the other station in different ways: by hugging the ground, by bouncing once or several times off the ionosphere, by going almost straight up and down for regional contacts, or through special effects such as sporadic E, aurora and meteors.
The wave follows the Earth's surface. Its range drops quickly with frequency and depends on the ground: on 160 m it covers 100–300 km (much more over seawater), on 80 m a few tens of kilometres, above 10 MHz only a few. Mainly useful on the low bands during the day.
In practice: Local and regional contacts on 160 m and 80 m during the day, from tens to a few hundred kilometres.
The ionosphere bends a signal launched at an angle back to Earth, hundreds to thousands of kilometres away. For every frequency above the critical frequency there is a steepest angle that still comes back; closer in lies the skip zone, which you cannot reach by sky wave — and often not by ground wave either.
In practice: Every DX contact on shortwave. A station 300 km away that cannot hear you while one 3000 km away can? It is in your skip zone.
Near Vertical Incidence Skywave: transmit almost straight up on a frequency below foF2 and the signal comes down in a circle of 0–400 km around you, with no skip zone. Ideal for regional contacts on 80, 60 and 40 m with a low horizontal dipole (0.1–0.25 λ high). It works only while foF2 is above your frequency — at night often only on 80 m.
In practice: Contacts 50–400 km away on 80 m or 40 m, for example a regional net. Hang your dipole low — a few metres is enough.
The MUF (maximum usable frequency) is the highest frequency the ionosphere returns on a given path. It is a median: on half of the days the band goes a little higher, on the other half not. The LUF (lowest usable frequency) is set by D-layer absorption and by your station — more power or a weak-signal mode lowers it. The best frequency, the FOT, is about 85 % of the MUF: open on roughly nine days out of ten.
In practice: Pick a band just below the MUF that HAMIOS shows. MUF 18 MHz? Then 20 m (14 MHz) is a good choice and 15 m (21 MHz) is closed.
Along the day/night boundary the D layer is weak on both sides — not yet formed on the sunrise side, already gone on the sunset side — while the F layer is still dense. Paths along this line, at sunrise and sunset, often give the best low-band DX of the day on 160, 80 and 40 m. HAMIOS draws the gray line from the Sun's exact position.
In practice: Around sunrise and sunset, look at the gray line on the HAMIOS map: stations on that line are the ones to try on 80 m and 40 m.
A signal can go the short way around the globe or the long way — the other side of the great circle, 40 000 km minus the short-path distance. When the short path is closed, or runs through darkness or the auroral zone, the long path through daylight or along the gray line is sometimes much stronger: Europe to New Zealand on 20 m in the morning is a classic.
In practice: Is a station weak? Turn your beam 180° and try the long path.
Dense Es clouds reflect 10 m, 6 m and occasionally 2 m over 500–2300 km, far above the normal MUF. Openings appear suddenly and can move quickly. Two hops (Es + Es, or Es + F2) occasionally connect continents on 6 m.
In practice: If 10 m or 6 m suddenly fills with stations 1000–2000 km away on a summer day, that is sporadic E — be quick, it can close again within minutes.
On both sides of the geomagnetic equator the ionosphere forms two dense crests (the equatorial anomaly). Paths crossing the equator roughly north–south — Europe to southern Africa, Japan to Australia — can carry 10 m and 6 m in the afternoon and evening, even when F2 elsewhere is closed. Best around the equinoxes.
In practice: In the afternoon and early evening, try 10 m towards southern Africa or South America.
During geomagnetic storms charged particles pour into the auroral zones. On HF that means extra absorption and a distorted ionosphere: polar paths (Europe–Japan, Europe–western North America) weaken first. On 6 m and 2 m the aurora itself can reflect signals, with a typical rough, buzzing sound — point your antenna north.
In practice: When Kp is high, avoid paths over the pole. On 6 m and 2 m, try pointing your antenna north.
Meteors leave short-lived trails of ionisation at 80–120 km that reflect VHF for a fraction of a second up to a minute. With MSK144, contacts of 800–2200 km on 6 m and 2 m are routine, especially during showers such as the Quadrantids (January), Perseids (August) and Geminids (December).
In practice: Use MSK144 during a meteor shower; the early morning hours are best.
Longer paths use several hops. Every ground reflection costs a few dB — less over sea than over dry land. On a chordal hop the signal skips from one ionospheric point to the next without touching the ground, which gives surprisingly strong signals on long paths, typically around the equator and in the evening.
In practice: The more hops, the weaker the signal — that is where CW and FT8 make the difference.
Whether a band “is open” is largely a question of geometry: the steeper a signal meets the layer, the lower the frequency the layer can still return. That single fact explains the skip zone, NVIS, and why a long DX path supports higher frequencies than a short one.
Think of skimming a stone on water: thrown flat it bounces, thrown steeply it sinks. A radio wave that meets the ionosphere at a flat angle is sent back even at high frequencies; one that meets it steeply comes back only at low frequencies. That is why a long path often works on a higher band than a short one — and why there is a zone around you that sky waves skip.
A wave that meets a layer at an angle of incidence φ is returned up to f = foF2 / cos φ — the secant law. The longer the hop, the flatter the incidence and the higher the maximum frequency. On a round Earth the factor for a 3000 km hop is about 3 to 3.5: that is the M(3000) factor ionosondes publish, and MUF(3000) = foF2 × M(3000).
| Take-off angle | F2 hop (300 km) | MUF / foF2 | E hop (110 km) | MUF / foE |
|---|---|---|---|---|
| 3° | 3,225 km | 3.3 × | 1,776 km | 5.2 × |
| 5° | 2,877 km | 3.2 × | 1,486 km | 4.9 × |
| 10° | 2,193 km | 2.9 × | 1,004 km | 4.0 × |
| 15° | 1,714 km | 2.6 × | 730 km | 3.2 × |
| 20° | 1,374 km | 2.3 × | 561 km | 2.6 × |
| 30° | 934 km | 1.8 × | 366 km | 1.9 × |
| 45° | 561 km | 1.4 × | 214 km | 1.4 × |
| 60° | 328 km | 1.1 × | 125 km | 1.1 × |
How to read it: a signal leaving your antenna at 10° comes down about 2200 km away after one F2 hop, and on that path the band can go up to 2.9 times the critical frequency. With foF2 = 7 MHz that is about 20 MHz — so 20 m works, 15 m does not. (Single hop on a round Earth, reflection at the given height.)
A horizontal dipole radiates most strongly at an angle set by its height above ground: at about λ/2 high the main lobe points about 30° up, at 1 λ about 14°. Low antennas (below λ/4) radiate mostly straight up — perfect for NVIS, poor for DX. Verticals radiate at low angles, but over average ground they lose the very lowest ones. The HAM Antenna Designer computes this pattern for your height and ground.
The ionosphere breathes with the Sun. Over a day the MUF rises after sunrise, peaks in the early afternoon and falls in the evening; the LUF follows the Sun even more closely. The window between the two curves is where the bands are open.
Sunlight charges the mirror. By day the higher bands (20–10 m) open; at night the lower bands (80–40 m) take over. Summer and winter, and the Sun's 11-year cycle, shift that picture.
Before sunrise the MUF is at its lowest and the DX is on 80–30 m. After sunrise the higher bands open one by one, while the D layer builds up and closes 160 m and 80 m for long distances. In the afternoon 20–10 m are at their best; after sunset the process reverses. The ionosphere lags the Sun by roughly an hour, so the MUF keeps rising a little after local noon.
Summer brings long days and sporadic E, but at mid-latitudes the daytime MUF is often lower than in winter: the summer upper atmosphere contains more molecules, which speed up recombination (the winter anomaly). Winter brings short days, long nights and the best low-band season. Around the equinoxes, in March and September, both hemispheres are lit alike — the best time for north–south DX and transequatorial propagation.
Every 11 years or so the Sun goes from quiet to active and back. Near solar maximum the daytime MUF can exceed 35 MHz and 10 m is open worldwide; near minimum 20 m is often the highest reliable DX band and 10 m opens mainly through sporadic E. Cycle 25 peaked in October 2024 (smoothed sunspot number 161) and is now in its declining phase — still good years for the high bands, but less often and less long each day.
HAMIOS shows a handful of numbers from the Sun and the magnetosphere. Here is what each one means for your bands.
Two kinds of numbers. The first tells how strongly the Sun charges the mirror: solar flux (SFI) and sunspots — higher is better for the high bands. The second tells how disturbed Earth's magnetic field is: Kp, Bz and the A-index — lower is better. Solar flares (the X-ray class) can knock out shortwave for minutes to hours.
Measured daily at 2800 MHz (10.7 cm) by the Dominion Radio Astrophysical Observatory in Penticton, Canada. The radio flux itself ionises nothing, but it tracks the solar EUV that creates the F2 layer: a higher SFI means a higher MUF. It runs from about 65 at solar minimum to 200–300 on the most active days.
The International Sunspot Number counts sunspot groups and individual spots (R = 10 × groups + spots) and is maintained by SILSO in Brussels. Averaged over a month or more it is the best single predictor of foF2 — which is why HAMIOS's model is built on it. Daily values jump around; for propagation the trend matters more.
Kp sums up geomagnetic disturbance over 3 hours on a scale from 0 (quiet) to 9 (extreme storm), from magnetometers around the world. HAMIOS uses NOAA's minute-by-minute estimate, so you see a storm while it develops. A storm hits HF twice: first extra absorption at high latitudes, then — often for a day or two — a lower foF2 and MUF (the negative storm phase).
The north–south component of the magnetic field carried by the solar wind, measured by DSCOVR and ACE at the L1 point, 1.5 million km sunward. When Bz turns south (negative) it couples to Earth's field and lets solar-wind energy in; a Bz of −10 nT or lower that lasts for hours drives storms. The solar wind needs 30–60 minutes from L1 to Earth, so Bz is the earliest warning you get. HAMIOS shows Bz live and raises an event when it is strongly southward.
Flares are classified by their peak X-ray flux: A, B, C, M and X, each ten times stronger than the previous. The X-rays arrive in about 8 minutes and suddenly boost the D layer on the sunlit side — a short-wave fade-out, lowest bands first. NOAA's R scale links the classes to their radio impact:
The daily A index sums up a whole day of geomagnetic activity. It is the average of the eight 3-hourly a indices — the linear equivalents of Kp (Kp 3 ≈ a 15, Kp 5 ≈ a 48). Below about 8 the day was quiet, 30–49 means a minor storm, 50 and above a major storm. A high A in the morning means the bands may still be recovering — high latitudes last.
NOAA sums up space weather in three scales, each from 1 (minor) to 5 (extreme). HAMIOS shows them as events in its advice.
| Scale | Caused by | Effect on HF |
|---|---|---|
| R1–R5 | Radio blackout: X-rays from a flare (M1 … X20) | Minutes to hours of fade-out on the sunlit side; low bands first |
| S1–S5 | Solar radiation storm: energetic protons (≥ 10 MeV) | Polar cap absorption: polar paths blacked out, often for days |
| G1–G5 | Geomagnetic storm: Kp 5 … 9 | Absorption at high latitudes, then a lower MUF for a day or more |
HAMIOS combines a physical model with live measurements: the model fills in what is not measured, the measurements keep the model honest.
HAMIOS works out what should be open, checks that against live measurements and real reports from other stations, and then tells you in plain words which band to use, which way to point and for how long.
Every panel — the band bars, the 24-hour heatmap, the advice and the propagation map — uses the same model. It starts from the sunspot number and the Sun's position, then calibrates itself on the nearest ionosonde's live foF2 and MUF (KC2G / GIRO network, refreshed every 15 minutes). Across 27 ionosondes worldwide the calibrated model matches the measured MUF with a median ratio of 1.00.
NOAA's D-RAP map shows where a flare or a proton event is absorbing HF right now. HAMIOS raises the LUF there, so after a flare the low bands drop out of the advice exactly where they are really closed.
Up to five recommendations — band, mode, frequency, direction and time window — checked against real WSPR, DX-cluster and PSKReporter spots. When real spots show a path the model does not expect (TEP, sporadic E), the spots win.
For the chosen band the map shows the probability that each point on Earth can be reached from your QTH, including the skip zone around you. Click anywhere for the great-circle path and the distance.
The day/night boundary and the gray line come from the Sun's exact position (NOAA solar calculator, within about 1° on the map). The aurora layer shows NOAA OVATION's measured probability of aurora, updated every few minutes.
The ionosonde panel shows your measured ionosphere: foF2 (the highest NVIS band) and MUF(3000) (the highest DX band). NOAA R / S / G scales, solar-wind shocks, southward Bz and measured sporadic E appear as events in the advice.
Time-stamped alerts for storm onset, strong flares, southward Bz, approaching lightning and satellite passes — each with a plain explanation of what it does to your bands.
HAMIOS keeps 90 days of solar and band data in charts, so you can see what a storm did to your bands or how the solar flux has been trending.
The models HAMIOS uses, exactly as they are in the source code (HAMIOS 5.8). They are deliberately simple and fast; the accuracy comes from calibrating them on live measurements.
This part is for the technically curious: the actual formulas behind HAMIOS. You do not need them to use the program — each one comes with a line in plain words.
In plain words: Where the Sun is, to within a hair — so HAMIOS knows exactly where it is day and where it is night.
Solar declination, equation of time and the subsolar point follow the NOAA solar calculator (after Meeus), accurate to about 0.01°. Day and night use a Sun elevation of −0.83° (refraction plus the Sun's radius). The same function drives the map, the gray line and the propagation model.
In plain words: How strong the mirror is: stronger with more sunspots and more sunlight; weaker at night, near the poles and during storms.
foF2 follows from the sunspot number (SSN) and the Sun's zenith angle one hour earlier — the ionosphere lags the Sun. Between a night value and a noon value the shape follows (cos z / cos z_noon)^0.6. It is reduced at high latitudes and during storms (Kp above 3).
In plain words: If a measuring station is nearby, HAMIOS trusts its measurement more than its own estimate — fully for now, less and less for later hours.
The nearest ionosonde within 2000 km, with a measurement younger than 90 minutes and a reliable ionogram (confidence ≥ 50), sets two correction factors. They count fully now and fade with a half-life of 8 hours for forecasts further ahead.
In plain words: How much of the signal is soaked up on the way: most around noon, hardly any at night. More power or FT8 lets you go a little lower.
The LUF grows with the Sun's elevation (D-layer absorption) and the solar flux, rises at high latitudes during storms and is lowered by your station's margin (mode, power, antenna). Where NOAA D-RAP measures absorption, the LUF is at least the highest affected frequency it reports.
In plain words: Turns the MUF and LUF into a simple percentage per band.
The MUF is a median: at the MUF a band is open on about half of the days, at 0.85 × MUF on about nine out of ten. Two logistic curves, one at each end of the window, give the chance per band.
In plain words: Long paths are checked at every bounce point, and the weakest point decides. Short paths get a lower maximum — that is the skip zone.
Paths up to 4000 km are evaluated at their midpoint, longer paths at ¼ and ¾ of the way (one control point per hop of about 3000 km); the weakest point decides. Short hops get a lower MUF because the signal meets the layer more steeply — that is what creates the skip zone on the propagation map.
In plain words: Where the northern and southern lights are — measured by NOAA, or estimated from Kp when that measurement is missing.
Normally the map shows NOAA OVATION's measured aurora probability. Without it, HAMIOS falls back to a Kp-driven oval around the IGRF-2025 geomagnetic poles:
In plain words: Where each satellite is — accurate to about 10 km, without extra software.
Satellite positions use Kepler's equation with the secular J2 perturbations (precession of the ascending node and of the argument of perigee) and the TLE's decay term — within about 10 km of the full SGP4 model over a day, without external libraries.
Known limits. The base model has no winter anomaly and no equatorial anomaly, and it treats the ionosphere over a path as smooth. Near an ionosonde the calibration corrects most of that; far from any ionosonde the model is on its own — which is exactly why HAMIOS lets real WSPR, DX-cluster and PSKReporter spots overrule it.
