Guide: how to read this site

Every panel carries two things: where the information came from (its classification) and how old it is (its freshness). Reading those two together tells you how much to trust a number.

  1. Check the classification — is it measured, forecast, modelled or derived?
  2. Check the freshness — is it live, cached or stale?
  3. Read the value itself (a band rating, a MUF, a reliability figure).
  4. For paths, remember the figure is a model, not a measurement of the ionosphere over your antenna.

Nothing on this site should be used for safety-of-life or critical communications. See the disclaimer.

Data classification

Every value is labelled with one of these four words.

LabelMeaningExample on this site
MEASUREDDirectly observed by an instrument. The most trustworthy class, but still limited by sensor and location.Kp index, solar flux, solar-wind speed
FORECASTA provider's prediction of a future value. Confidence falls off with distance in time.Predicted F10.7, predicted Kp, R/S/G outlook
MODELLEDOutput of a propagation model using current conditions — a calculation, not an observation.Path MUF and reliability
DERIVEDThis project's own calculation built on top of other data, such as a band rating or an overall outlook.Best Bands Now, HF Outlook

Freshness states

Shown as a pill or in the Source Status panel, with the age in minutes.

StateMeaning
LIVEFresh upstream data, collected recently (currently within about 20 minutes).
CACHEDThe newest data we hold is older than usual, or the provider is temporarily unreachable and we are showing the last good snapshot.
STALEData older than the configured threshold. It is still shown, but must not be read as current.
UNAVAILABLENo valid data. The panel will say so rather than inventing a value.

Cached data is never presented as live: the original observation timestamp is always displayed next to it.

Band ratings

Ratings in Best Bands Now and the HF Outlook are DERIVED by this project from the current solar and geomagnetic state plus the time of day.

RatingWhat it suggests
EXCELLENTThe band is well supported right now; expect strong, reliable signals and good DX potential.
GOODUsable and dependable, with normal fading. Good chance of contacts.
FAIRMarginal. Signals may be weak, fluttery or short-lived; openings can be brief.
POORLargely unsupported. Occasional weak openings possible.
CLOSEDNo useful propagation expected on this band at this time.

A rating is a general expectation for the Cape Town area, not a guarantee for your exact location, antenna or power.

Badges and tags

BadgeMeaning
DERIVEDCalculated by this project from other data (see classification).
NOAA FORECASTProvider forecast data, from NOAA SWPC. Public domain; attribution given.
ITU-R P.533The path predictions come from the official ITU-R P.533-14 implementation (ITURHFProp). This is the international standard method for HF circuit prediction.
PHASE 1 MODELShown instead of the P.533 badge only if the ITU engine is unavailable on the server. A simplified fallback calculation, clearly labelled so it is never mistaken for the full model.
LIVE LAYERSEverything drawn on the map is computed from current data: daylight, grey line and paths.
PHASE DRemaining planned work. Currently this means the modelled MUF heat overlay on the global map; the underlying P.533 engine is already running.

Propagation model and paths

MUF — Maximum Usable Frequency
The highest frequency that the ionosphere can support for a given path at a given time. Above the MUF, signals pass through into space instead of returning to Earth. A path MUF of 14.8 MHz means 20 m (14 MHz) is likely usable and higher bands probably are not.
Reliability
An estimate of how often a circuit at a given frequency would meet a usable signal-to-noise level, expressed as a percentage. Higher is more dependable. It is a statistical figure from a model, not a measurement.
Path
A circuit between two points — here, always starting at Cape Town. Domestic paths stay inside South Africa; DX paths leave the region.
Great-circle path
The shortest route over the Earth's surface between two points, drawn as a curve on the map. Radio follows this route, not the straight line on a flat map.
ITU-R P.533-14
The international standard method for predicting HF circuit performance. This site runs the official ITU-R implementation (ITURHFProp), using the current month, UTC hour and sunspot number, and an ITU reference antenna (type 141, a centre-fed antenna with an aperiodic screen reflector). Outputs used here are the basic MUF and the basic circuit reliability.
Phase D
Remaining planned work — currently the modelled MUF heat overlay on the global map. The P.533 engine itself is already in use.

Map layers

Daylight / Night shading
The dark area is where the Sun is below the horizon right now. HF behaves very differently on the sunlit and dark sides of the Earth: the D layer absorbs lower bands in daylight, while the F layer supports higher frequencies.
Grey line
The twilight band along the day/night boundary (the terminator). Paths that run along the grey line can enjoy unusually good DX, because both ends of the path are in near-darkness while the D layer has already weakened. The map draws the grey line explicitly.
Subsolar point
The single point on Earth where the Sun is directly overhead at that moment. The map marks it, and the whole day/night pattern follows from it.
Path colour
Green, amber or red, reflecting the modelled reliability of that path: green is dependable, amber is marginal, red is unlikely.

Space-weather terms

SFI / F10.7 — Solar Flux
The 10.7 cm radio flux, a proxy for solar activity and ionisation. Roughly: below 100 is low, 100–150 is normal, above 150 is high. Higher flux generally supports higher HF bands.
Kp — Planetary K index
A 0–9 measure of geomagnetic disturbance, in thirds (0, 3⅓, 6⅔ …). Quiet is 0–3, unsettled around 4, storm from 5 upward. Higher Kp depresses the MUF and hurts high-latitude paths most.
Ap
A daily linear version of the K index, used for longer-term averages.
Sunspots
The observed sunspot number, a long-standing indicator of solar cycle activity. Higher generally means more ionisation and better high-band conditions.
X-ray class
Background X-ray flux from the Sun, classified A, B, C, M, X in increasing order of intensity. C-class flares can cause minor fadeouts; M and X can cause significant shortwave fadeouts on the sunlit side.
R / S / G scales
NOAA's three space-weather scales: R radio blackouts, S solar radiation storms, G geomagnetic storms. Each runs 1–5 (minor to extreme). 0 means no event.
Solar wind speed
The speed of the charged particle stream from the Sun, in km/s. Around 300–400 km/s is calm; sustained speeds above about 500 km/s often precede disturbed conditions.
Bz / IMF
The north–south component of the interplanetary magnetic field, in nanotesla (nT). A sustained southward Bz (negative) couples energy into the magnetosphere and is a strong warning of coming geomagnetic disturbance.

Space-weather alerts

NOAA issues coded messages. This site decodes them for you.

Message code
The NOAA product code, for example ALTK04. The first three letters give the type: ALT = alert, WAR = warning, WAT = watch, SUM = summary, EXT = extended. The rest identifies the phenomenon and serial.
ALERT / WARNING / WATCH
An alert means the threshold has been reached. A warning means conditions are expected or in progress. A watch means conditions are possible. Alerts are the most immediate.
Threshold reached / Synoptic period
When the measured value crossed the threshold, and the three-hour observation window it belongs to.
NOAA scale (R/S/G)
Where an event maps onto the R/S/G scales, the site shows the scale and its plain-language effect on HF.

Time

SAST
South African Standard Time, UTC+2, with no daylight saving. The local clock on this site.
UTC
Coordinated Universal Time, the reference used across radio, space weather and logging. Both SAST and UTC are shown so logs and bulletins match.
Time base
The site's clock is taken from a GPS-disciplined NTP time source, so the timestamps and data ages shown are referenced to an accurate standard rather than to your device's clock.

South African specifics

ZS call areas
South African amateur callsigns are numbered by region: ZS1 Western Cape, ZS2 Eastern Cape, ZS3 Northern Cape, ZS4 Free State, ZS5 KwaZulu-Natal, ZS6 Gauteng, Mpumalanga, North West and Limpopo. The domestic path table uses these areas as destinations.
Ionosonde
A radar that measures the ionosphere directly by sounding it vertically. South Africa's network is operated through SANSA at sites including Hermanus and Grahamstown.
foF2 / hmF2 / T-index
Ionosonde outputs: foF2 the critical frequency of the F2 layer (which sets the ceiling for vertical propagation), hmF2 its height, and the T-index a smoothed measure of ionospheric activity.
Where ionosonde data comes from
South African ionosonde observations (foF2 and hmF2) are used with written permission from the South African National Space Agency (SANSA) and are displayed as published on SANSA's public Space Weather site, with attribution and a link back. Only the published values are shown; SANSA is acknowledged as the data provider and does not endorse this site.

How the predictions are verified

Model output is only useful if it can be shown to be right. Four independent checks are used, in increasing order of strength.

1. Reference regression (automated)
The upstream ITU-R project publishes golden input/output pairs taken from its own validation set (test case IDs 1 and 164). The engine installed on this server is re-run against those inputs and the numeric results are compared line by line with the published reference output. They match exactly. If the engine were mis-compiled, reading the wrong data files, or mis-configured, this test would fail.
2. Parameter pass-through (automated)
Confirms that the year, month, hour and sunspot number this site sends are actually received and echoed by the model, so a silently-ignored input field cannot go unnoticed.
3. Geometry consistency
The great-circle distance this site computes for each path is compared with the distance the models report for the same circuit. Cape Town to Johannesburg: this site 1261.6 km, ITU-R P.533 1261.6 km, VOACAP 1261.8 km.
4. Independent cross-check (automated)
Two unrelated model families — ITU-R P.533 and NTIA VOACAP — are run over the same paths with the same inputs and power, and their band recommendations are compared. Agreement is reported per path; disagreement is shown rather than hidden.

These checks run automatically on the server with every model run.

What verification does not mean

These checks show the implementation is faithful and self-consistent. They do not prove that any individual prediction will match what you hear on the air. Propagation forecasting is statistical: a “90% reliable” circuit is expected to be usable for about 90% of the month, not every minute of it. The strongest available test — comparing predictions against real reception reports or ionosonde measurements — is the natural next step and is described on this site's roadmap.

Acronyms

TermMeaning
HFHigh Frequency — 3 to 30 MHz
MUFMaximum Usable Frequency
SNRSignal-to-Noise Ratio
NVISNear Vertical Incidence Skywave — short-range HF using steep angles
DXLong-distance contacts
IMFInterplanetary Magnetic Field
SWPCSpace Weather Prediction Center (NOAA)
P.533ITU-R Recommendation for HF circuit prediction
VOACAPVoice of America Coverage Analysis Program (NTIA/ITS)
SANSASouth African National Space Agency
POPIAProtection of Personal Information Act (South Africa)

Data sources and their licensing are listed on the Data Sources page and the Licensing page.