Ionospheric structure effects on HF radio wave propagation for the Enhanced Polar Outflow Probe (e-POP) satellite mission



[1] The Enhanced Polar Outflow Probe (e-POP) payload will be launched on a small satellite in 2007 for exploring plasma and atmospheric outflow process in the polar region. The subject of this paper is whether one can determine the properties of large-scale ionospheric structures by studying the perturbations that they cause on HF radio waves received at the e-POP satellite from ground-based transmitters such as the Canadian Advanced Digital Ionosonde (CADI). The perturbations on the received waves have been investigated using numerical ray-tracing methods. These simulation results show that ionospheric irregular structures lead to a complex pattern of amplitude, propagation time delay, Doppler frequency, and direction-of-arrival (DOA) effects on the HF radio waves received at the satellite. The simulations also show that ionospheric density structure cannot be measured unambiguously using a single wave property. Therefore a “catalog” of HF signatures of typical ionospheric irregularities has been established in order to be able to interpret the e-POP HF measurements in terms of ionospheric structures.

1. Introduction

[2] The Enhanced Polar Outflow Probe (e-POP) is a small-satellite payload that will be used to study the atmosphere and plasma flows and related wave particle interaction processes in the high-latitude region of the ionosphere. This mission was originally proposed to the Canadian Space Agency (CSA) in 1996 and the planned launch date of the satellite is in 2007 with a 12-month operation period. e-POP will perform in-situ observations in selected high-latitude portions of its orbit and do coordinated operations with ground radar facilities such as the Canadian Advanced Digital Ionosonde (CADI) [MacDougall et al., 1995], and the Super Dual Auroral Radar Network (SuperDARN) [Greenwald et al., 1995]. The e-POP satellite will be in a polar, elliptical orbit with an inclination between 70° and 90°, a perigee between 300 and 400 km and an apogee of 1500 km. As the first space environment orbital payload to be developed in Canada since 1971, e-POP will provide a unique opportunity for the renewal of expertise in space science instrument development in Canadian research institutions [Yau et al., 2002].

[3] As early as the 1940s, scientists had already realized that extraterrestrial radio sources could be utilized to study the ionosphere [Hey et al., 1946]. From the beginning of the artificial-earth-satellite era (Sputnik I, 1957), beacon satellites have been used to study ionospheric irregularities. The amplitude, phase, direction-of-arrival (DOA) and frequency of a radio wave traversing the ionosphere are distorted when the signal encounters drifting or stationary irregularities. These distortions, or “scintillations”, when measured with ground receivers, provide information about the structures and motions of the irregularities [Hunsucker, 1991]. More recently, the GPS satellite constellations and satellites in the Global Navigation Satellite System (GLONASS) have been used to study the ionospheric morphology and irregularities using TEC and tomographic measurements [Pi et al., 1997; Kersley et al., 1997; Dabas and Kersley, 2003; Leitinger et al., 1997]. These observations use signals from satellite transmitters at VHF and higher frequencies. However, lower HF frequencies are more sensitive for probing ionospheric density structures. Extraterrestrial sources of HF radiation that are detectable on the ground, principally the Sun and Jupiter, are of limited use for ionospheric imaging except for such techniques as imaging riometers [Detrick and Rosenberg, 1990]. In the e-POP experiment the usual positions of transmitter and receiver will be reversed: the e-POP satellite will contain the receiver, and the transmitters will be on the ground. The e-POP payload includes a Radio Receiver Instrument (RRI) to measure the amplitude, DOA, group delay, and Doppler shift of signals transmitted from ground stations [James, 2003]. These HF radio wave measurements will be coordinated with simultaneous observations at ground facilities.

[4] The present paper describes simulation studies of HF radio wave propagation from a stationary ground transmitter to the e-POP satellite using ray-tracing methods. To date, there has been little exploitation of the idea of using a known source on the ground and a coordinated spaceborne receiver to image structure within the ionosphere. The aim of this investigation was: (a) to determine characteristics of signals received at the satellite by ray tracing simulations through a number of ionospheric models containing large scale irregularities, (b) to study the relationship between signal behavior patterns and the parameters of the irregularities, and (c) to establish an understanding of the effects of HF propagation through irregularities in order to set transmitter and receiver parameters for the e-POP radio experiments.

2. Ionospheric Irregularity Models

[5] The existence of F region irregularities has been known for at least 40 years from their effects on transionospheric radio propagation, but our knowledge of them is still incomplete [Hunsucker and Hargreaves, 2003]. The irregularities can be classified into three groups according to size, (a) small-scale (<100 m); (b) intermediate-scale (0.1–10 km) (c) large-scale (>10 km) [Vickrey and Kelley, 1983]. Small and intermediate-scale irregularities cause phase and amplitude scintillations by a diffraction mechanism [Burns and Hargreaves, 1996; Hunsucker and Hargreaves, 2003]. In this study we are mainly concerned with large irregularity structures. These typically have 50–1000 km horizontal scales and can have remarkably high plasma density. Large scale enhancements within the polar cap regions are usually called “patches”, while in the auroral or subauroral zones they are commonly called “blobs”. The polar cap patches appear when the interplanetary magnetic field (IMF) is southward with plasma concentration in the patches often more than a factor of two higher than the background level (sometimes the density increase can be as much as ten times) [Crowley, 1996; Hunsucker and Hargreaves, 2003]. Such structures were first reported by Weber et al. [1985], and later establishment of polar observatories by different institutes identified these features as being present in both summer and winter [McEwen et al., 1995; Smith et al., 2000].

[6] Bank et al. [1974] made early observations of blobs in the auroral zone using the Chatanika incoherent scatter radar. Other observations at Chatanika and at EISCAT have been reported in many papers [Vickrey et al., 1980; Muldrew and Vickrey, 1982; Tsunoda et al., 1985; Hargreaves and Burns, 1988]. There are three types of blobs based on their location: subauroral blobs, boundary blobs, and auroral blobs. The boundary blobs are located at the equatorward boundary of the auroral zone and have extreme longitudinal extent and persist for very long times. They are often collocated with the poleward wall of the midlatitude ionospheric trough [Crowley, 1996]. The midlatitude trough (sometimes called the “main” trough) is a feature in the subauroral ionosphere [Moffett and Quegan, 1983; Roger et al., 1992]. This is a region of depleted ionization, latitudinally limited in width and extended in the east-west direction. Within the trough, critical frequencies are typically reduced to below half the value of those outside the trough region. Experimental observations and simulation studies of the effects of the trough on HF long propagation paths have been undertaken by many authors [Lockwood, 1981; Stocker et al., 2003; Warrington and Stocker, 2003]. Helms and Thompson [1973] investigated the HF propagation through a trough using ray-tracing simulations and focused on ground-based observations using frequencies lower than the critical frequency. HF radio propagation from ground to a satellite passing above structures such as the trough has not been previously of interest so has not been described in the literature.

[7] For the present study we therefore required models of patches, blobs and the trough. For the trough we based our model (see Figure 1) on the measurements shown by Weber et al. [1985]. In their paper, Weber et al. presented a time-history of irregular structures observed by the Chatanika incoherent scatter radar (65.1°N, 147.5°W) on January 29, 1979. Major features of the high-latitude ionosphere are shown in their Figure 2c for the first scan (from 0543UT to 0557UT) along the magnetic meridian: (a) a stratified F-layer enhancement that extends from low latitudes to just north of Chatanika (at zero location); (b) a “boundary blob” just north of the Chatanika; (c) a F-layer trough separating the previous two features; (d) an auroral E layer that extends equatorward of the boundary blob and into the region of the F layer trough. We model it as shown in Figure 1 and refer to this as “Model Trough” in this paper. The background electron density started with a typical ionosphere based on the International Reference Ionosphere (IRI), but for mathematical convenience we actually used a cosine layer below, and a Chapman layer above the height of hmF2, and set key parameters such as NmF2 and hmF2 using the IRI2001 model on the same selected day and time as the Weber et al. measurements. The vertical dashed line in Figure 1 indicates the center location of the trough region.

Figure 1.

Irregular Model Trough. Trough and enhancements modeled from a Chatanika incoherent scatter radar scan (65.1°N, 147.5°W) on January 29, 1979 [Weber et al., 1985].

Figure 2.

Amplitude fluctuations caused by Model Trough. Radio frequency 5.5 MHz. The x-axis is the horizontal distance from south to north; x = 0 is the location of the transmitter. The vertical line shows the center of the trough region (see Figure 1), and the dash-dotted curve shows the signal strength for a horizontally stratified ionosphere.

[8] To produce the simple patch and blob irregularity models, we multiplied the entire electron density profile by an enhancement factor with a Gaussian shape as a function of horizontal distance. Several irregularities (Gaussian distribution) with different dimensions and maximum enhancement respect to the referenced ionospheric profile were used in these studies, and in this paper we will be referring to results using three of these which we call Model 1, 2, 3 (see Table 1). They have similar shapes to the “blob” region north of the trough in Figure 1.

Table 1. Irregularity Models and Their Maximum Deviation With Respect to the Background and the Width of Gaussian Disturbance Between the Points Half of the Maximum Deviation
Irregularity ModelsMaximum Deviation, %Width of Gaussian Disturbance, km
Model 1+20100
Model 2+2025
Model 3+40100
Model Trough−5550

[9] For our simulations we assumed that the ground transmitter was at Saskatoon (52.16°N, 106.5°W), and that the e-POP satellite passed from south to north at an altitude of 850 km with a circular orbit in the magnetic meridian. The ray-tracing program integrates the Haselgrove differential equations [Haselgrove, 1963] using the Runge-Kutta method, and uses the Appleton-Hartree formula as the index of refraction for normalization [Yabroff, 1961].

3. Simulated Example Measurements and Discussions

[10] Ray trajectories of HF radio waves are significantly altered in the vicinity of an electron density trough or enhancement. This phenomenon is more obvious when the radio wave frequency is chosen to be just above the critical frequency of the layer. With a magnetic field imposed, the ray paths for the ordinary mode are found to display patterns that are similar to those for the extraordinary mode, except that the extraordinary rays are reflected slightly lower in the ionosphere, propagate over shorter ground distances and refract slightly equatorward. Ray-tracing results also showed that elimination of the small E region near 100 km height and from 0 ∼ 200 km north in Model Trough did not lead to significant changes in the ray patterns. This finding agrees with Helms and Thompson's [1973] results.

[11] As a wave travels through an irregular medium, the fluctuation of amplitude and phase is a cumulative process along each ray path. The measured amplitude of a signal depends on many factors such as: (a) the traveling distance (b) power of the transmitted signal (c) frequency (d) refraction (e) polarization of the signal (f) direction of the arrival (g) absorption and scattering losses (h) radiation patterns of antennas at both transmitter and receiver sides (i) interference effects of multi-paths of the propagation [Mass, 1962]. These diverse influences makes the computation of amplitude rather complicated. In our simulation study, the signal strength was calculated from the power density at the altitude of the satellite orbit with the transmission power 600 W, a uniform radiation pattern for the satellite receiving antenna with an effective length of 3 m. Multi-path effects were effectively included by simply determining the average power that the satellite would receive from all the ray paths that would enter a ‘box’ of length 25 km along the satellite orbit.

[12] Figure 2 (panels 1–6) shows average power (converted to amplitude μV) of the received signal due to the presence of Model Trough at various locations. For comparison the signal amplitude for a horizontally unstructured (reference) electron density profile is shown by the dash-dotted line in each of the panels. The radio frequency, 5.5 MHz, was chosen to be slightly higher than the critical frequency of the “blob” north of the trough region. The vertical dashed line shows the center location of the trough region relative to the transmitter. These amplitude patterns show both large-scale features and small scale (fast) fluctuations. The large-scale features are very dominant when the trough is nearly overhead the transmitter, and there are deep signal depletions bordered with strong signal enhancements. When the trough region is displaced more than 200 km from the transmitter the large-scale features become much less dominant, and the fast fluctuation areas move to distant e-POP receiver locations. It should be noted that the energy concentration may be two or more times higher than the maximum amplitude of the received signal for the reference electron density profile when the trough structure is nearly overhead.

[13] Figure 3 shows the amplitude deviation with respect to the reference profile for different single large-scale enhancement models (Models 1 to 3) located overhead. A comparison of the results for Models 1 and 3 in the figure shows that there is an increase of amplitude fluctuations with increase of irregularity electron density. This figure also shows, by comparison of the amplitude deviation for Models 1 and 2, that smaller irregularities such as Model 2 affect a smaller area but give larger fluctuations. These results for simple irregularity structures can be visualized as the imperfectly focused patterns of ionospheric lenses.

Figure 3.

Signal amplitude deviation, with respect to amplitude variation for the reference ionosphere, for three different simple ionospheric structure models. Irregularity models are centered over the transmitter.

[14] As discussed above, the group paths and the propagation delay time of radio waves fluctuate in the presence of irregular regions. In our simulations we also computed propagation delays and compared these delays with the normalized propagation time delays corresponding to the horizontally stratified reference ionosphere. Figure 4 shows the normalized group delay in the presence of the Model Trough located at various positions. If the trough region is in the vicinity of the transmitter (Figure 4, panels 2–4) the propagation time is shorter compared to the reference delays for the central part of the satellite pass. There are also increased delays particularly when the satellite is north of the trough. The amount of delay increase depends on the density of the irregular enhancement, so rays traversing the higher density region north of the trough show larger delay from which it may be possible to estimate the density of the boundary blob.

Figure 4.

Normalized group delay in the presence of trough Model Trough. Group delay = 1 is the group delay for the reference ionosphere. The vertical line shows the center position of the trough region relative to the transmitter. If the trough region is in the vicinity of the transmitter (zero position), there is a negative deviation (see panels 2 and 3).

[15] Another important effect of irregularity structure is to cause perturbations in the direction-of-arrival (DOA) of the radio waves arriving at the e-POP satellite receiver. By measuring the phases of the received signal on a set of monopole receiving antennas it is possible to determine the DOA at the satellite. The DOA of the received signal at the satellite is a geometrical factor, and depends on the satellite position and refractive effect of the ionosphere. We show in Figure 5 the DOA difference between an ionosphere with irregularity Model Trough and the reference profile. Comparison of Figure 2 with Figure 5 shows similar locations for significant fluctuations of DOA and significant variations of signal amplitude. When an electron density enhancement becomes the dominant feature affecting the propagation, such as the boundary blob, the DOA deviation increases with increase of the density of the enhancement. The DOA fluctuations are more significant when the irregularity is nearly overhead. The fluctuation is smaller when the irregularity is north or south of the transmitter.

Figure 5.

DOA deviation for Model Trough. DOA = 0 is the DOA of the reference ionosphere. The vertical line represents center of the trough region.

[16] We next show (Figure 6) the DOA perturbations due to simple irregularity structures. For Figure 6 our Model 1 was used for the irregularity. In this figure (top row) we also show the amplitude variations. The vertical line in each panel marks the center of the irregularity. If this simple irregularity is located overhead of the transmitter (panels 1 and 4), the signal amplitude pattern as a function of satellite position is nearly symmetrical (panel 1) and the DOA pattern (panel 4) is antisymmetric. As the irregularity moves away from being over the transmitter (panels 2, 3 and 5, 6), both the amplitude and DOA patterns quickly loose their symmetry and become just complex variations, particularly for amplitude.

Figure 6.

Signal amplitude and DOA variations for a simple ionospheric structure (Model 1).

[17] A further property of the radio waves that can be measured by e-POP is the Doppler shift. From simulations, the effect of ionospheric irregularities on Doppler frequency shift is observed to produce relatively complicated patterns. Some Doppler shift results are shown in Figure 7. Doppler shift does not seem to be a very useful measurement for determination of the properties of large scale ionospheric structures.

Figure 7.

Doppler frequency shift, radio frequency:11 MHz. The irregularity models are centered over the transmitter.

[18] Our simulations show that there is no single radio wave measurement by e-POP/RRI that would unambiguously determine the irregularity properties. Each irregularity produces a complex pattern for the various measured radio wave properties. However, by comparing a measured pattern with our simulations patterns it should be possible to determine the probable irregularity structure.

[19] An additional objective of this study was to determine the required RRI parameters for irregularity studies. A summary of a large number of our simulation results and consequent requirements for the RRI specifications are given in Table 2.

Table 2. Predicted Wave Parameters and RRI Specification
Predicted Wave Parameters in Typical Large-Scale IrregularitiesRRI Measurement Specification
NameRange of Variation Over All Models
Amplitudemin: 0 μV0–2000 μV
max: 1500 μV
Delay2.0–10 ms0–15 ms GPS based accuracy: ±8 μs
DOAAll elevations above horizontal and all azimuthsAccuracy of DOA determination: ±0.5°
Doppler0–1000 HzAccuracy of ±10 Hz

4. Conclusions

[20] The future e-POP mission will provide an exciting tool for probing the ionosphere in the polar region. This paper investigates propagation processes between a ground-based transmitter and the e-POP satellite using ray-tracing simulations. The irregularities lead to diverse propagation patterns, especially when the radio frequency is close to the ionospheric critical frequency. Angular redistribution of radiation energy flux accompanies the fluctuation of DOA and displays significant effects if rays encounter irregularity boundaries where there are sharp gradients in electron density. Rays with higher elevation angles are affected more than rays with lower elevation angles. Subject to the propagation and scattering geometry, the propagation time may be either decreased or increased in relation to its reference value in the absence of irregularities. The magnitude of this effect depends on the irregularity density [Zabotin et al., 2001]. The simulations show that a decreased group delay time is the best indicator of the existence of a trough region. The delay deviation is a function of the location of the irregularity relative to the transmitter and shifts in location as the irregularity is displaced from above the transmitter.

[21] Our analysis shows that large-scale ionospheric structures would produce very observable effects on HF radio wave propagation from a ground transmitter to the satellite. We have established the magnitudes of the received wave variations due to typical irregularities that e-POP may encounter. However, any single measurement cannot be used to determine unambiguously the ionospheric density structure. By comparing the e-POP measurements with a library of radio wave variations from our simulations, it should be possible to determine the basic properties of the structures.


[22] This work was supported by a grant from Natural Science and Engineering Research Council of Canada.