Understanding Useful NVIS Receive Angles

Which elevation angles should an NVIS receive antenna cover? “Straight up” is too narrow an answer, but replacing it with one universal high-angle window is not much better. Start with the coverage area and the current propagation conditions. The sky decides which rays are usable; the installed antenna and local noise decide which of them produce the best SNR.

NVIS Is a Path Family, Not One Arrival Angle

NVIS uses steep ionospheric paths to cover the region beyond reliable ground-wave service and inside the skip zone of lower-angle skywave modes. The required launch and arrival elevation changes with ground distance. A station almost underneath the returned ray can use energy close to vertical; a station farther from the midpoint generally uses a lower elevation for the same effective ionospheric height.

The distribution also changes when the effective returning region moves, when E- and F-region modes coexist, or when the ordinary and extraordinary magneto-ionic components follow different group paths. Terrain, ground wave and other skywave modes can overlap the received signal. Calling the result “the NVIS angle” hides all of that structure.

Virtual-Height Geometry Is a Planning Sketch

A useful first drawing treats a one-hop path as two straight segments meeting at a virtual height h′ above the midpoint. For a short path on a flat Earth, with total ground distance d:

Here e is the terminal elevation angle. If the same illustrative virtual height is 250 km, a 100 km path gives about 79°, while a 300 km path gives about 59°. Those numbers are not recommended coverage limits. They demonstrate why one height cannot produce one angle for every regional circuit.

An ionogram’s virtual height is inferred from group delay. NOAA’s NCEI documentation explicitly distinguishes it from true physical height because a pulse slows while propagating through the ionized medium. The real ray refracts continuously; Earth curvature, horizontal gradients, magnetic field, several layers and several modes require ray tracing or an HF propagation model. Use the simple equation to catch impossible assumptions, not to certify an antenna.

Frequency Determines Whether the Path Exists

A useful receive angle does not matter if the chosen frequency is not supported by the current path. The upper boundary is path- and time-dependent. ITU-R distinguishes a basic MUF from an operational MUF: the latter includes the terminals, antennas, emission and required performance. Near vertical incidence, the vertical critical frequencies measured by an ionosonde are important inputs, but an oblique path MUF is not simply a band label or one nearby foF2 reading.

The lower boundary is also operational. D-region absorption rises on some lower-frequency daytime paths, atmospheric noise can dominate the lower HF and MF range, and solar flares or energetic-particle events can sharply increase absorption. “Below MUF” is therefore necessary but not sufficient. The useful frequency must also deliver the required SNR and reliability above absorption, noise and interference.

Choose ionosondes near useful control regions, inspect the ionogram and confidence flags rather than copying one auto-scaled number, and compare more than one station when horizontal gradients matter. A climatological model such as IRI is valuable for planning; NASA describes IRI as a monthly-average empirical standard model. It becomes more representative when appropriate measured foF2 or hmF2 inputs constrain it, but it is still not an arrival-angle measurement.

Band Names Do Not Assign Elevation Angles

One regional path may work on 40 metres in daytime and require a lower frequency after sunset. Another location or solar state may not support the same 40-metre path at all. A low-frequency path may be limited by absorption and noise even while reflection is physically possible. On 160 metres, ground wave and skywave can also overlap in ways that make “NVIS-only” interpretation unsafe.

For each available amateur allocation, run the same path and time analysis. Record the predicted modes and elevation distribution, then compare them with observed signal, noise and fading. Do not transfer a 40-metre angle to 80 metres or a winter-night result to a summer afternoon.

The Installed Pattern Matters More Than the Antenna Name

A horizontal dipole, inverted V, doublet, loop or compact receiving antenna does not come with one NVIS pattern. Height in wavelengths, conductor geometry, slope, ground conductivity and permittivity, loss, nearby structures, feedline routing and common-mode current all change the realized pattern. In a multiband installation, the same physical height becomes a different electrical height on every band.

Model the full installed geometry over a bounded set of plausible ground parameters. Inspect realized gain—not directivity alone—through the complete elevation and azimuth region required by the paths. Look for nulls and strong azimuth variation, and include mismatch and system loss. If feedline-exterior current is significant, the feedline and station are part of the receiving structure until that path is controlled.

The peer-reviewed measurements by Witvliet and colleagues are useful precisely because they separate transmit field strength from receive SNR and state their farmland ground case, path, frequency and antenna geometry. Their reported optima are evidence for those cases, not a metre-height prescription for another soil, antenna or noise environment.

Maximum Signal Is Not Maximum SNR

For a passive linear antenna in the same environment, reciprocity links its transmit and receive directional response. The best receive installation can nevertheless differ from the best transmit installation because reception is an SNR problem. Wanted skywave, atmospheric noise, distant interference and local man-made noise do not have to share an elevation, azimuth, polarization or coupling path.

ITU-R P.372 treats atmospheric, galactic and man-made noise as distinct contributions. Its scope explicitly excludes noise that reaches the receiver through other conductors or inadequate feeder balance, so those installed coupling paths must be measured separately. A lower broadband noise indication is useful only when receiver bandwidth, attenuation, preamplifier, AGC and detector are controlled and the wanted signal is recorded at the same time.

Polarization Diversity Is Conditional

The magnetized ionosphere supports ordinary and extraordinary characteristic waves with different phase, group delay, absorption and polarization behaviour. Their relative amplitude and phase can vary, so a signal launched with linear polarization may arrive with changing elliptical polarization. Multipath and mode interference add amplitude and selective fading.

Two orthogonal antennas, separated antennas or frequency-diverse channels can reduce an outage only when their fades or noise are sufficiently decorrelated and the combining method preserves SNR. Shared local noise, pattern nulls, receiver overload, unequal gain or a badly calibrated combiner can erase the benefit. Diversity is a measured system result, not a guarantee attached to crossed elements or an active probe.

A Measurement-Led NVIS Receive Workflow

• Define the service. List the coverage points, time windows, frequencies, modes, required SNR and acceptable outage probability.
• Collect propagation inputs. Inspect nearby and path-relevant ionograms, foF2 and height parameters with quality flags; add current absorption and space-weather context.
• Predict a distribution. Use a declared P.533/P.1240 implementation or ray tracer to obtain candidate modes and elevation angles across the path, frequency, season, hour and ionospheric variability—not one median ray.
• Model the installation. Include complete antenna geometry, height in wavelengths, realistic ground cases, nearby conductors, feedline loss and exterior-current paths.
• Commission with references. Log several stable regional transmitters or coordinated test signals with fixed receiver settings; record signal, noise, SNR, fading and weather for every antenna state.
• Change one variable. Compare height, geometry, termination, choke or polarization with rapid A/B/A switching or simultaneous channels.
• Repeat over time. Check daytime and night-time, several seasons and disturbed conditions. Update the usable frequency and angle set when the ionosphere changes.

Primary and Authoritative References

• ITU-R P.533-14 — Prediction of HF circuit performance, modes, elevation angles and field strength
• ITU-R P.1240-2 — Basic/operational MUF and ray-path prediction
• ITU-R P.1239-4 — Reference ionospheric characteristics, foF2, M(3000)F2 and virtual height
• ITU-R P.373-8 — Operational MUF, basic MUF, OWF and LUF definitions
• ITU-R P.372-17 — Atmospheric, galactic and man-made radio noise
• NOAA NCEI — Vertical ionograms, critical frequency and virtual-height interpretation
• NOAA NCEI — Real-time ionosonde data, foF2 and hmF2 products
• NOAA SWPC — Current space-weather and D-region absorption products
• NASA CCMC — International Reference Ionosphere 2020 model, inputs and statistical scope
• Witvliet et al. — Measured NVIS elevation angles and case-bounded receive/transmit antenna height
• Witvliet and Alsina-Pagès — Peer-reviewed NVIS propagation, antenna, diversity and channel overview
• Witvliet and Alsina-Pagès — Magneto-ionic NVIS propagation measurement probes and sensors

Joeri’s Bottom Line

Do not replace the cartoon of a 90° ray with another rigid angle window. Define the regional paths, find the frequencies that the current ionosphere can support, predict their high-elevation distribution, and make the installed antenna cover that distribution without an important null.

Then measure signal and noise separately. The most useful NVIS receive geometry is the one that provides repeatable SNR and reliability for your paths and times—not the one that wins a universal height or angle table.

Questions, antenna-factor records or height trials to share? Contact RF.Guru.