Atmospheric, Ionospheric, and Satellite Propagation

Introduction and Context

Radio propagation is determined not only by the transmitting and receiving antennas but also by the physical medium between them. The atmosphere is electrically non-uniform, time-varying, and dispersive. Its effects range from small changes in phase and path length to complete signal blockage through absorption, scintillation, or ionospheric reflection. These effects are especially important in HF communication, satellite links, radar, navigation systems, remote sensing, and deep-space communication.

Propagation analysis is conventionally divided into three regions:

1. Tropospheric propagation, involving the lowest atmospheric layer and extending approximately to 8–18 km altitude.
2. Ionospheric propagation, involving the ionized upper atmosphere, approximately from 60 km to beyond 1,000 km.
3. Satellite propagation, involving an Earth-space path through the atmosphere and free space.

These regions overlap physically, but the dominant mechanisms differ. The troposphere produces refraction, ducting, attenuation, rain scattering, and gaseous absorption. The ionosphere produces dispersion, refraction, reflection, Faraday rotation, absorption, and scintillation. Satellite links experience free-space spreading loss together with atmospheric impairments and geometric constraints.

Learning Objectives

After completing this chapter, the reader should be able to:

• explain the atmospheric structure relevant to radio propagation;
• distinguish phase velocity, group velocity, refraction, diffraction, scattering, and ducting;
• describe ionospheric layers and the physical basis of HF propagation;
• calculate critical frequency, maximum usable frequency, skip distance, free-space loss, and link-budget quantities under stated assumptions;
• evaluate rain, gaseous, ionospheric, and scintillation impairments;
• design and assess a practical satellite link;
• interpret propagation measurements and apply engineering mitigation techniques;
• recognize regulatory, operational, and research issues in modern propagation systems.

1. The Propagation Medium

For a plane electromagnetic wave in a homogeneous, lossless medium, the phase velocity is

where v_p is phase velocity in m/s, c is the speed of light in vacuum (approximately 3 × 10⁸ m/s), and n is refractive index. In the atmosphere, n is close to unity but varies with pressure, temperature, and water-vapour content.

The refractive index is commonly expressed through refractivity:

For radio frequencies below approximately 100 GHz, a widely used approximation for standard atmospheric refractivity is

where P is total atmospheric pressure in hPa, e is water-vapour partial pressure in hPa, and T is absolute temperature in kelvin. This empirical expression assumes ordinary atmospheric conditions and is not a complete model for all frequencies or extreme environments.

A useful modified refractivity is

where h is height in metres and M is expressed in N-units. The term involving height accounts approximately for Earth curvature. A sufficiently negative vertical gradient of M indicates conditions favourable for anomalous propagation or ducting.

Refraction and ray curvature

Snell’s law for a stratified medium is

where θ₁ and θ₂ are angles measured from the normal at an idealized interface. In a continuously varying atmosphere, rays bend gradually toward regions of higher refractive index. Since refractive index generally decreases with altitude, ordinary rays bend slightly toward Earth, extending the radio horizon.

The effective-Earth-radius approximation replaces the actual curved ray by a straight ray over an enlarged Earth. Under standard conditions, the effective Earth radius is approximately 4/3 of the physical radius. The radio horizon for transmitter and receiver heights h_t and h_r, in metres, is approximately

where a is Earth’s mean radius, approximately 6.37 × 10⁶ m, and k is the effective-Earth-radius factor. This approximation applies to smooth terrain and gradual atmospheric gradients; mountains, buildings, ducting, and strong weather gradients can make it inaccurate.

Diffraction and scattering

Diffraction permits energy to reach regions beyond a geometrical line of sight, particularly around terrain obstacles. Its importance increases as wavelength increases and obstacle dimensions become comparable with the Fresnel-zone dimensions. Scattering results from irregularities in refractive index or from particles such as raindrops, ice crystals, and dust. Unlike specular reflection, scattering distributes energy over a range of directions and can produce both useful over-the-horizon links and unwanted fading.

2. Tropospheric Propagation

The troposphere contains most atmospheric water vapour and nearly all weather systems. Its influence is particularly strong above approximately 1 GHz, where precipitation and gaseous absorption become increasingly important.

2.1 Standard and anomalous propagation

Under standard atmospheric conditions, the refractive index decreases with height at a rate that produces ordinary downward ray bending. When temperature and humidity profiles produce a stronger-than-normal decrease, rays may bend sufficiently to follow the Earth over long distances. This is called super-refraction. If rays become trapped between atmospheric layers or between the Earth and an elevated inversion layer, ducting occurs.

Ducts may be surface-based or elevated. They can produce unexpectedly strong signals far beyond the normal radio horizon, but they can also create interference between distant systems using the same frequency. Ducting is common over oceans and during temperature inversions.

2.2 Atmospheric attenuation

Atmospheric attenuation consists mainly of:

• oxygen and water-vapour absorption;
• rain attenuation;
• cloud and fog attenuation;
• scattering by hydrometeors;
• depolarization and differential attenuation.

The received power is often represented in decibels. For independent loss mechanisms,

where each term is a positive attenuation in dB. The terms should not be added blindly when they are correlated or when a propagation model already includes one impairment within another.

For a uniform rain path, a common engineering model is

where A_R is rain attenuation in dB, γ_R is specific attenuation in dB/km, and L_eff is effective rain-path length in km. A power-law approximation for specific attenuation is

where R is rain rate in mm/h and k and α depend on frequency, polarization, temperature, and drop orientation. This is a statistical model: a selected rain rate corresponds to a specified percentage of time, not a guaranteed worst case.

Horizontal polarization generally suffers greater rain attenuation than vertical polarization at many microwave frequencies because raindrops become flattened as they fall. Circular polarization can reduce some polarization-alignment problems but does not eliminate attenuation or depolarization.

2.3 Atmospheric noise

Atmospheric noise arises from lightning, thunderstorms, solar activity, and thermal emission from gases and precipitation. At lower frequencies, lightning-generated noise can dominate receiver noise in rural and tropical environments. At microwave frequencies, atmospheric emission contributes to the system noise temperature and is especially important for low-elevation satellite links.

3. Ionospheric Propagation

The ionosphere is a partially ionized plasma produced mainly by solar ultraviolet and X-ray radiation. Its electron density varies with altitude, latitude, local time, season, solar cycle, geomagnetic activity, and sporadic events.

The main regions are:

These altitude ranges are approximate and vary substantially.

3.1 Plasma frequency and critical frequency

In an unmagnetized, collisionless plasma, the plasma angular frequency is

and the plasma frequency is

when N_e is expressed in electrons per cubic metre and f_p is in hertz. Here e is electron charge, m_e is electron mass, and ε₀ is free-space permittivity.

For vertical incidence, the critical frequency of an ionospheric layer is approximately the plasma frequency corresponding to its maximum electron density:

The numerical coefficient assumes N_max in m⁻³. A wave with frequency below the critical frequency can be refracted back toward Earth under the simplified vertical-incidence model. A wave above it generally penetrates the layer, although oblique incidence changes the condition.

3.2 Oblique incidence and MUF

The secant law provides an idealized relation between critical frequency and maximum usable frequency:

where θ_i is the angle of incidence measured from the vertical at the ionospheric layer. Equivalently, if ψ is measured from the layer normal, the same convention must be used consistently. The secant law assumes a horizontally stratified, smoothly varying ionosphere and neglects magnetic-field effects, collisions, absorption, and irregularities.

The maximum usable frequency is the highest frequency expected to support a specified path with an acceptable reliability. In practice, operators use a frequency below the predicted MUF, often called the frequency of optimum transmission, to provide a margin against temporal and spatial variation.

3.3 Skip distance and skip zone

The skip distance is the minimum ground distance at which a particular frequency returns to Earth after ionospheric refraction. Between the end of the ground-wave service area and the first sky-wave return may be a skip zone, where reception is weak or absent. The exact geometry depends on layer height, electron-density profile, take-off angle, Earth curvature, and multiple-hop propagation.

A simplified flat-Earth single-hop geometry with virtual reflection height h and incidence angle θ from the vertical gives

This is only a geometric approximation; it should not be used as a precise prediction for long paths or strongly curved rays.

3.4 Ionospheric absorption and magnetoionic effects

Electron-neutral collisions, especially in the D region, convert electromagnetic energy into heat. Absorption increases at lower frequencies and is generally greater during daytime. Solar flares can produce sudden ionospheric disturbances that increase HF absorption and cause radio blackouts.

The Earth’s magnetic field makes the ionosphere anisotropic. A wave can separate into ordinary and extraordinary magnetoionic modes, each with different phase velocities and absorption. The resulting differential phase shift rotates the plane of polarization, an effect known as Faraday rotation. It is important for satellite communication, remote sensing, and transionospheric links.

The ionosphere is also dispersive. Group delay is approximately inversely proportional to the square of frequency. A first-order approximation for ionospheric group delay is

where Δt is delay in seconds, TEC is total electron content in electrons per square metre, and f is carrier frequency in hertz. The expression assumes a thin, cold, non-magnetized plasma and a suitable slant-path TEC. At lower frequencies and for high-precision work, higher-order geomagnetic terms and ray-path effects must be included.

3.5 Scintillation

Ionospheric scintillation is rapid fluctuation of received amplitude and phase caused by small-scale electron-density irregularities. It is particularly severe near the equatorial anomaly and at high latitudes, often after sunset or during geomagnetic disturbances. Spread-F, equatorial bubbles, and traveling ionospheric disturbances are important sources.

Mitigation includes link margin, spatial diversity, frequency diversity, adaptive coding and modulation, interleaving, robust synchronization, and multi-frequency correction. No single method is optimal for all satellite navigation or communication systems.

4. Satellite Propagation Geometry and Link Budget

A satellite link consists of an uplink, a space segment, and a downlink. The basic link budget is an accounting of transmitted power, antenna gains, propagation losses, and receiver losses.

For free-space propagation, the power received by matched antennas is given by the Friis transmission equation:

where P_t and P_r are transmitted and received powers in watts, G_t and G_r are dimensionless antenna gains in the direction of the path, λ is wavelength in metres, and R is separation in metres. The assumptions are free space, far-field operation, matched polarization, no obstruction, and a narrowband signal.

The corresponding free-space path loss is

or in decibels,

where frequency is in GHz and distance is in km. The constant 92.45 is valid for these units. It is essential not to combine this expression with a wavelength form while retaining the same numerical constant.

A practical link budget in dB is

The terms L_atm, L_pol, and L_point represent atmospheric, polarization, and pointing losses. Receiver performance is commonly expressed through carrier-to-noise density ratio:

where k_B is Boltzmann’s constant and T_sys is system noise temperature in kelvin. In dB-Hz, 10 log₁₀(k_B T_sys) is the thermal noise spectral density in dBW/Hz.

For a data system of bit rate R_b and required energy-per-bit to noise-density ratio E_b/N₀,

where R_b is in bit/s. This relation assumes that C represents the useful information carrier power and that implementation losses are included separately.

Worked Example: A simplified geostationary downlink

A ground station receives a 12 GHz signal from a satellite at a slant range of 38,000 km. The satellite effective isotropic radiated power is 50 dBW, the receiving antenna gain is 44 dBi, and total atmospheric, polarization, pointing, and miscellaneous losses are 3 dB. Determine the received power.

Using the free-space-loss equation:

Since 20 log₁₀(12) ≈ 21.58 dB and 20 log₁₀(38000) ≈ 91.60 dB,

The received power is therefore

This is an idealized clear-sky link-budget result. Real design must additionally consider availability percentage, rain-rate statistics, antenna pointing, polarization mismatch, feeder loss, interference, satellite output back-off, and receiver implementation margin.

5. Satellite Orbits and Propagation Consequences

A geostationary satellite appears fixed over the equator because its orbital period equals Earth’s sidereal rotation period. Its principal engineering advantages are continuous coverage and fixed ground antennas. Its disadvantages include long propagation delay, large free-space loss, and poor elevation angles at high latitudes.

Low-Earth-orbit satellites operate at much lower altitude and therefore provide lower path loss and delay, but the satellite moves rapidly relative to the ground. They require tracking, handover, constellation coordination, and Doppler compensation.

Medium-Earth-orbit systems provide a compromise between coverage, delay, constellation size, and path loss. Orbit selection is therefore a system-level trade-off rather than merely an antenna decision.

Propagation delay is approximately

for a one-way path of length R. A complete communication interaction may include uplink, downlink, processing, terrestrial routing, and return paths; hence application latency is greater than the single free-space propagation delay.

6. Applied Case: Designing a Tropical Ka-Band Satellite Link

Background

Consider an illustrative fixed satellite service operating near 20 GHz downlink in a tropical coastal region. The service must provide 99.5% annual availability to a hub station and distributed terminals.

Problem

The initial clear-sky link budget closes with 6 dB margin. However, intense tropical rainfall produces several-minute outages during storms. The operator must improve availability without simply increasing satellite transmit power.

Analysis

At 20 GHz, rain attenuation is a major impairment. A point rainfall rate is not identical to path attenuation because rain intensity varies along the slant path. The engineering process therefore uses a rain-rate distribution, an estimated rain height, slant-path geometry, horizontal reduction factors, and a target percentage of time. Polarization-dependent attenuation and depolarization are also assessed.

The link is most vulnerable at low elevation angles because the slant path through rain is longer. Site diversity may be effective if two gateways are separated by a distance sufficient to make their intense-rain events weakly correlated.

Approach and solution

The operator adopts four measures:

1. adaptive coding and modulation, reducing information rate during deep fades;
2. uplink power control, within satellite power and interference constraints;
3. a geographically separated backup gateway;
4. a larger terminal antenna and improved pointing calibration.

A propagation-monitoring system records beacon amplitude, local rain rate, elevation angle, and modulation state. These measurements validate the statistical attenuation model and allow maintenance teams to distinguish rain fades from hardware faults.

Outcome

The combined solution improves service availability more efficiently than a fixed power increase. Adaptive operation protects basic service during moderate attenuation, while site diversity handles severe localized storms. The larger antenna improves clear-sky margin but does not remove the need for rain statistics or operational control.

Learning points

• Frequency selection must be evaluated together with availability requirements.
• A link budget is a probability-based design, not a single deterministic calculation.
• Diversity and adaptation can be more cost-effective than excessive transmitter power.
• Local propagation measurements are essential when climate models are uncertain.

7. Engineering Design and Mitigation Practices

Reliable systems begin with a propagation requirement stated quantitatively: frequency range, path geometry, availability, fade margin, data rate, latency, polarization, and permitted interference. Good practice includes:

• use ITU-R propagation recommendations and locally validated climate data;
• calculate clear-sky and impaired link budgets separately;
• include antenna gain degradation, feeder loss, radome loss, and pointing error;
• reserve margin for model uncertainty and equipment aging;
• use site, frequency, polarization, or time diversity where appropriate;
• monitor beacons and environmental sensors;
• apply adaptive coding and modulation only with stable synchronization and well-defined control limits;
• coordinate spectrum use and protect adjacent satellite and terrestrial systems;
• verify models through long-term measurements rather than short demonstrations.

Artificial intelligence is increasingly used to forecast rain fade, classify scintillation, estimate channel state, and select modulation modes. However, machine-learning predictions must be trained on representative local data, tested against physically based models, and monitored for distribution shift. A black-box prediction should not replace safety margins, spectrum coordination, or engineering validation.

8. Limitations, Regulation, and Research Directions

Propagation models contain uncertainty because atmospheric and ionospheric conditions are stochastic and spatially non-uniform. A model calibrated in a temperate region may perform poorly in equatorial, polar, desert, or maritime environments. Resolution is also a limitation: a weather radar, numerical weather model, and link measurement observe different spatial scales.

Satellite systems operate within national and international regulatory frameworks. Frequency allocation, orbital-slot coordination, equivalent isotropically radiated power limits, out-of-band emissions, antenna sidelobe requirements, and interference protection are governed through national administrations and international coordination, especially under the International Telecommunication Union framework. Engineers must treat regulatory compliance as part of system design rather than a final administrative step.

Important contemporary research areas include:

• high-frequency broadband links above Ka band, where rain becomes increasingly severe;
• integrated satellite–terrestrial networks;
• optical satellite links, which avoid radio-frequency congestion but suffer from clouds, atmospheric turbulence, and pointing constraints;
• improved ionospheric specification using GNSS networks and distributed sensors;
• reconfigurable antennas and electronically steered arrays;
• propagation-aware resource allocation for non-geostationary constellations;
• physics-informed machine learning for short-term channel prediction.

Chapter Summary

Atmospheric, ionospheric, and satellite propagation determine the performance, availability, and design margin of radio systems. Atmospheric refractivity changes ray paths and can produce ordinary refraction, super-refraction, ducting, and anomalous interference. Tropospheric attenuation is caused primarily by gases, clouds, rain, and scattering, with rain becoming especially important at microwave and millimetre-wave frequencies.

The ionosphere is a variable plasma whose electron density controls plasma frequency, critical frequency, maximum usable frequency, absorption, dispersion, Faraday rotation, and scintillation. HF systems exploit ionospheric refraction, while satellite navigation and communication systems must often mitigate ionospheric delay and irregularities.

Satellite links combine free-space spreading loss with atmospheric, polarization, pointing, and equipment losses. The Friis equation and decibel link budget provide the basic analytical framework, while C/N₀ and E_b/N₀ connect received power to communication quality. Practical design requires statistical propagation models, availability targets, diversity, adaptive transmission, monitoring, and regulatory coordination.

Key Terms

Refractive index: Ratio of electromagnetic phase velocity in vacuum to that in a medium.

Refractivity: Quantity N = (n − 1) × 10⁶ used to express small atmospheric deviations of refractive index.

Super-refraction: Atmospheric refraction stronger than the standard condition.

Ducting: Trapping of radio energy within an atmospheric layer or between a layer and Earth.

Troposphere: Lowest atmospheric region containing most weather and water vapour.

Ionosphere: Ionized upper-atmospheric region affecting radio-wave propagation.

Plasma frequency: Natural oscillation frequency of electrons in a plasma.

Critical frequency: Maximum vertical-incidence frequency reflected by an ionospheric layer under an idealized model.

Maximum usable frequency: Highest frequency predicted to support a specified ionospheric path and reliability.

Skip distance: Minimum ground distance for a specified sky-wave return.

Skip zone: Region between ground-wave coverage and the first sky-wave return.

Total electron content: Integrated electron density along a propagation path.

Faraday rotation: Rotation of polarization caused by magnetoionic propagation.

Scintillation: Rapid amplitude and phase fluctuation caused by small-scale irregularities.

Free-space path loss: Power loss associated with spherical spreading in unobstructed space.

Link budget: Accounting of gains, losses, and received power in a communication path.

Rain attenuation: Signal loss caused by absorption and scattering in precipitation.

Site diversity: Use of geographically separated terminals to reduce correlated fades.

C/N₀: Carrier power divided by noise spectral density, expressed in dB-Hz.

Availability: Percentage of time a link meets its specified performance requirement.

Review Questions

Short Answer Questions

1. Define atmospheric refractivity and explain why it is useful in propagation analysis.
2. What is the difference between ordinary refraction, super-refraction, and ducting?
3. Define critical frequency and maximum usable frequency.
4. State two important causes of satellite-link attenuation other than free-space loss.
5. What physical phenomenon produces ionospheric scintillation?

Long Answer Questions

1. Explain the structure of the ionosphere and discuss the propagation roles of the D, E, and F regions.
2. Derive the conceptual basis of the free-space link budget from the Friis transmission equation.
3. Discuss rain attenuation at microwave frequencies, including statistical modelling and mitigation.
4. Explain ionospheric dispersion, total electron content, and Faraday rotation in transionospheric links.
5. Compare geostationary, medium-Earth-orbit, and low-Earth-orbit satellite systems from a propagation perspective.

Analytical / Application Questions

1. A 14 GHz satellite link has a slant range of 40,000 km. Calculate its free-space path loss using the dB formula and state the unit assumptions.
2. A layer has maximum electron density N_max = 1.0 × 10¹2 electrons/m³. Estimate its vertical critical frequency using the plasma-frequency approximation and discuss why the result is not a complete operational MUF prediction.
3. An uplink has 4 dB clear-sky margin but a predicted 7 dB rain attenuation at the target availability percentage. Propose a technically justified combination of mitigation measures and explain the trade-offs.

Multiple Choice Questions with Answers

Question 1

Which atmospheric quantity is directly related to the refractive index by N = (n − 1) × 10⁶?

A. Atmospheric refractivity

B. Total electron content

C. Rain attenuation

D. Critical frequency

Correct Answer: A. Atmospheric refractivity

Question 2

Under standard assumptions, which condition most strongly favours atmospheric ducting?

A. A sufficiently strong vertical gradient of modified refractivity

B. A uniform refractive index at all heights

C. Complete absence of water vapour

D. A perfectly conducting ground surface

Correct Answer: A. A sufficiently strong vertical gradient of modified refractivity

Question 3

The ionospheric plasma frequency primarily depends on which quantity?

A. Electron density

B. Rainfall intensity

C. Satellite altitude alone

D. Antenna aperture only

Correct Answer: A. Electron density

Question 4

For a vertical-incidence ionospheric wave, the critical frequency is approximately the frequency corresponding to the layer’s:

A. Maximum electron density

B. Minimum atmospheric pressure

C. Maximum rain rate

D. Ground-wave field strength

Correct Answer: A. Maximum electron density

Question 5

In the simplified secant law, the maximum usable frequency increases when the incidence angle from the vertical:

A. Increases, because sec θ increases

B. Decreases to zero, because sec θ decreases

C. Has no effect under any condition

D. Becomes imaginary

Correct Answer: A. Increases, because sec θ increases

Question 6

Which expression gives free-space path loss in terms of frequency in GHz and distance in km?

A. L_fs = 92.45 + 20 log₁₀(f_GHz) + 20 log₁₀(R_km)

B. L_fs = 20 log₁₀(f_GHz/R_km)

C. L_fs = 4πR/λ

D. L_fs = k_B T_sys

Correct Answer: A. L_fs = 92.45 + 20 log₁₀(f_GHz) + 20 log₁₀(R_km)

Question 7

A link has P_t = 40 dBW, transmit gain 30 dBi, receive gain 40 dBi, free-space loss 200 dB, and other losses 5 dB. What is the received power?

A. −135 dBW

B. −125 dBW

C. −130 dBW

D. −95 dBW

Correct Answer: A. −135 dBW

Question 8

Which impairment is especially important for many Ka-band satellite links?

A. Rain attenuation

B. Ground-wave diffraction only

C. Audio-frequency crosstalk only

D. Thermal noise elimination

Correct Answer: A. Rain attenuation

Question 9

Ionospheric group delay in the first-order approximation varies approximately as:

A. TEC/f²

B. TEC × f²

C. f/TEC

D. 1/(TEC × f²)

Correct Answer: A. TEC/f²

Question 10

Which mitigation technique uses geographically separated terminals to reduce correlated weather fades?

A. Site diversity

B. Polarization rotation only

C. Increasing symbol rate only

D. Removing link margin

Correct Answer: A. Site diversity