NASA Heliophysics Summer School – Boulder, CO, USA – 19 July 2013. Planetary and Solar Radio Occultation/Scintillations: With an Emphasis on Interplanetary Scintillation (IPS) and Heliospheric Faraday Rotation (FR). Dr. Mario M. Bisi (Institute of Mathematics and Physics, Aberystwyth University, Wales, UK) – Mario.Bisi@aber.ac.uk.
Lecture Outline ? A Brief Introduction to the Radio Spectrum! A Brief Introduction to the Radio Spectrum! ? ? What is Occultation? What is Occultation? Planetary Occultation Measurements. ? ? Planetary Occultation Measurements. ? ? What is Scintillation? What is Scintillation? ? ? Interplanetary Scintillation (IPS) Telescopes/Arrays. Interplanetary Scintillation (IPS) Telescopes/Arrays. An Introduction to IPS and Basic IPS Theory. ? ? An Introduction to IPS and Basic IPS Theory. ? ? Some Recent IPS Results and LOFAR Technical Advances. Some Recent IPS Results and LOFAR Technical Advances. A Brief Introduction to IPS Three- -Dimensional (3 Tomographic Reconstructions and Examples of their Usage. Tomographic Reconstructions and Examples of their Usage. ? Dimensional (3- -D) D) ? A Brief Introduction to IPS Three ? ? A Brief Introduction to Heliospheric Faraday Rotation (FR) A Brief Introduction to Heliospheric Faraday Rotation (FR) and Initial Progress Towards Heliospheric FR Determination. ? and Initial Progress Towards Heliospheric FR Determination. ? Lecture Summary. Lecture Summary. ?
A Brief Introduction to the Radio Spectrum!
The Electromagnetic (EM) Spectrum ? An overview of the EM spectrum; adapted from Heliophysics Volume 2, Chapter 4, by Tim Bastian. ? Commercial AM radio band lies in the LF-MF range, and the FM band lies in the VHF part of the spectrum.
The Radio Spectrum Frequency Band Definition Frequency Range Wavelength Range Band Designations ELF - Extremely Low Frequency ULF - Ultra Low Frequency VLF - Very Low Frequency LF - Low Frequency MF - Medium Frequency HF - High Frequency VHF - Very High Frequency < 300 Hz 300 Hz - 3 kHz 3 kHz - 30 kHz 30 kHz - 300 kHz 300 kHz - 3 MHz 3 MHz - 30 MHz 30 MHz - 300 MHz > 1,000 km 1,000 km - 100 km 100 km - 10 km 10 km - 1 km 1 km - 100 m 100 m - 10 m 10 m - 1 m - - - - - - - P (sometimes) = 300MHz - 1 GHz L = 1 GHz - 2 GHz S = 2 GHz - 4 GHz C = 4 GHz - 8 GHz X = 8 GHz - 12 GHz Ku - 12 GHz - 18 GHz K = 18 GHz - 26 GHz Ka = 26 GHz - 40 GHz Q = 30 GHz - 50 GHz U = 40 GHz - 60 GHz V = 50 GHz - 75 GHz E = 60 GHz - 90 GHz W = 75 GHz - 110 GHz F = 90 GHz - 140 GHz D = 110 GHz - 170 GHz UHF - Ultra High Frequency 300 MHz - 3 GHz 1 m - 100 mm SHF - Super High Frequency 3 GHz - 30 GHz 100 mm - 10 mm EHF - Extremely High Frequency 30 GHz - 300 GHz 10 mm - 1 mm ? An overview of the radio spectrum by band, frequency range, wavelength range, and band classification.
What is Occultation? (Who can tell me?)
Solar Occultation During a Full Eclipse ? An occultation occurs when one object is hidden by another that passes between it and the observer; i.e. when an apparently larger body passes in front of an apparently smaller one. ? The total solar eclipse of 2006 clearly displaying the faint solar corona in all its glory. Image taken from Bisi, Ph.D. Thesis, 2006; courtesy of Dr. Richard A. Fallows (currently at ASTRON in The Netherlands).
Planetary Occultation Measurements.
Jupiter’s Ionosphere ? The classical Doppler shift in signal Taken from Hinson et al., 1998. Measuring Jupiter’s Ionosphere. radio frequency was measured which was caused by the refractive bending in Jupiter’s atmosphere/ ionosphere.
Jupiter and Io ? According to the literature, Io is acted upon by a J x B force which tends to try to propel it out of the Jovian system. In addition, its motion about Jupiter puts it through the Io plasma torus at Jupiter which carries two billion kilowatts of power into the Jovian ionosphere. The energy source for these processes is from Jupiter's relatively-fast rotation. This is an unusual planet-satellite coupling. ? Figure: The Io flux tube (IFT) and its associated decametric emission cones – to scale (Belcher, 1987). ? See Belcher, 1987 for further information.
Jupiter and Callisto ? Electron density profiles were obtained at Callisto using data from the Galileo spacecraft with the radio occultation technique. There were five occultations by Callisto providing eight usable observing opportunities (entrance and exit observations). ? The detection of an ionospheres which were calculated as being generated by the photo-ionisation and electron-impact ionisation of the neutral gasses at the surface. ? Detectable ionospheres were only when the side in which the moon is travelling (the ramside) was in sunlight which suggests the need for both photo-ionisation and the impact of a plasma onto the upper atmosphere of the moon. ? Results were obtained from the Galileo spacecraft; see Kliore et al., 2002 for further information.
Saturn’s Ionosphere and Atmosphere radio occultation measurements of the Pioneer 11 flyby of Saturn on 01 ? From September ionosphere 1979, and the upper atmosphere were measured. ? Revealed that the primary electron density peak of the ionosphere was at a height of 1,800 km with a density around 11,400 cm-3. Saturn’s electron density between equatorial radii of 60,000 km and 70,000 km. The solid curve is From entry data into occultation and the ? For further information, see Kliore, et al., 1980 (figure dashed curve is a profile obtained from exit data by using an artificial drift function; not to be also taken from here). used for magnitudes of electron density.
Saturn and Titan, and Venus’ Ionosphere ? Saturn’s largest moon, Titan, occulted a bright giant K-type star designated 28 Sgr on 03 July 1989. ? Strong scintillations were recorded in the stellar signal throughout the occultation which are attributed to the refractive-index changes in Titan’s atmosphere (e.g. Hubbard et al., 1988). ? An extensive discussion on this stellar occultation is given by Hubbard et al., 1993. ? Initial observations of the night-side ionosphere of Venus were possible through radio-occultation measurements with the Pioneer Venus Orbiter Spacecraft. See Kliore et al., 1979. ? What about the Earth…
Earth’s Atmosphere: GPS Radio Occultations ? Allows for the derivation of atmospheric profiles. From Kursinski et al., 1997. ? Momentum and energy within the atmosphere is transported by small-scale waves which contribute to the middle atmospheric structure and circulation – the height is important. ? Radio occultation observations provide a high vertical resolution measure and sensitivity to density and temperature perturbations (from waves) and turbulence used to characterise properties in the atmospheres of Venus, Jupiter, Titan, Uranus, and Neptune. See Kursinski et al., 1997 and references therein, for further details.
What is Scintillation? (Again, who can tell me?)
Scintillation and Quasars ? George Gamow, “Quasar”, 1964. Twinkle, twinkle quasi-star Biggest puzzle from afar How unlike the other ones Brighter than a billion suns. Twinkle, twinkle quasi-star How I wonder what you are. ? Quasar 3C 273 as imaged by the Hubble Space Telescope's Advanced Camera for Surveys. Image is courtesy of NASA, A. Martel (JHU), H. Ford (JHU), M. Clampin (STScI), G. Hartig (STScI), G. Illingworth (UCO/Lick Observatory), the ACS Science Team and ESA.
Interplanetary Scintillation (IPS) Telescopes/Arrays
EISCAT, ESR, and MERLIN (224 MHz-~6GHz) Above: The European Incoherent SCATter radar (EISCAT) and EISCAT Svalbard Radar (ESR) radio telescopes from left-to-right: Tromsø, Norway (M.M. Bisi, October 2003); Kiruna, Sweden (M.M. Bisi, May 2003); Sodankylä, Finland (http://www.eiscat.com/sodan.html); and the ESR 42m in the foreground and steerable 32m in the background (M.M. Bisi, May 2005). Left: The Multi-Element Radio- Linked Interferometer Network (MERLIN) MkIa (Lovell) radio telescope at Jodrell Bank (near Manchester, England); and Right: The MERLIN MkII radio telescope also at Jodrell Bank (M.M. Bisi, May 2004).
The LOw Frequency ARray (LOFAR) (1) LOFAR Core High-Band Antenna (top) and LOFAR Core Low-Band Antenna (bottom); both with Dr. Richard A. Fallows (~ 5’ 5½” tall) in for size comparison. LOFAR superterp (top) and LOFAR Chilbolton (bottom).
LOFAR (2) LOFAR core in The Netherlands with stations around The Netherlands and International Stations in Germany (5), France (1), Sweden (1) and in the UK (1). The stations shown in green are complete and operational while yellow depicts stations that are under construction as of March 2013.
LOFAR (3) ? Frequency agile system over two primary observing bands of 10 MHz to ~240 MHz split into 10 MHz to 90 MHz (LBA) and 110 MHz to ~240 MHz (HBA) with two antenna types. ? Ample collecting area with plenty of combinations of multi-site observations with the International Stations over long baselines. ? Experimentation with beam modes to enable band widths encompassing 80 MHz to ~the entire observing frequency range! ? Possibilities of ~5° angular resolution in three-dimensional (3-D) tomographic reconstructions look plausible with LOFAR data. ? The Murchison Widefield Array (MWA) in Western Australia could match (or possibly exceed) the number of observations per day but will not offer the multi-site observations of LOFAR. ? Kilpisjärvi Atmospheric Imaging Receiver Array (KAIRA) based on LOFAR technology located just inside Finland.
Japan, India, and other IPS Arrays/Telescopes The Solar Terrestrial Environment Laboratory (STELab) antennas of Fuji (top left), Sugadaira (top middle), (new) Toyokawa (top right), (old) Toyokawa (bottom left), and Kiso (bottom middle); and the Ootacamund (Ooty) Radio Telescope (ORT) (bottom right) (Courtesy of http://stesun5.stelab.nagoya-u.ac.jp/uhf_ant-e.html, B.V. Jackson, and P.K. Manoharan). Others also include: MEXART, Mexico; Pushchino, Russia; UTR-2, Ukraine; and the Murchison Widefield Array (MWA), Australia.
An Introduction to IPS and Basic IPS Theory.
An Introduction to IPS (1) Radio signals received at each site are very similar except for a small time-lag. The cross-correlation function can be used to infer the solar wind velocity(s) across the line of sight (LOS). (Not to scale) Hubble Deep Field – HST (WFPC2) 15/01/96 – Courtesy of R. Williams and the HDF Team and NASA IPS is most-sensitive at and around the P-Point of the LOS to the Sun and is only sensitive to the component of flow that is perpendicular to the LOS; it is variation in intensity of astronomical radio sources on timescales of ~0.1s to ~10s that is observed.
An Introduction to IPS (2)
An Introduction to IPS (3) Density Turbulence ? Scintillation index, m, is a measure of level of turbulence. ? Normalized Scintillation index, g = m(R) / <m(R)>. • g > 1 → • g ≈ ≈ ≈ ≈ 1 → • g < 1 → → → → enhancement in δ δ δ δNe. → → → ambient level of δ δ δ δNe. → → → rarefaction in δ δ δ δNe. (Courtesy of Prof. P.K. Manoharan.) Scintillation enhancement with respect to the ambient wind identifies the presence of a region of increased turbulence/density and possible CME along the line-of-sight to the radio source.
An Introduction to IPS (4) Low-pass filter “Fresnel Knee” (speed) Power-Law (density spectrum) High-pass filter An example power spectrum from an observation of IPS with its key features marked. - Plot courtesy of Dr. Richard A. Fallows, ASTRON, The Netherlands.
An Introduction to IPS (5) Bpar= 75 km Bpar= 150 km Bpar= 210 km Klinglesmith, 1997 ? The ability to distinguish between streams of different velocity improves as the parallel baseline length (Bpar) increases between two observing sites; if (Bpar) is long enough, streams with different velocities appear as widely-separated peaks in the (temporal) cross-correlation function. ? The height of the maximum cross correlation decreases as parallel baseline length increases since density pattern changes with time.
Basic IPS Theory - Hewish, A., “A user's guide to scintillation”, Journal of Atmospheric and Terrestrial Physics, 51, pp.743-750, 1989. Abstract: Scintillation methods and the theory underlying scintillation methods are reviewed. Consideration is given to diffractive and refractive scintillation and describing irregular media. Application of scintillation are discussed, including ionospheric, interplanetary, and interstellar scintillation and scintillation and source size.
Basic IPS Theory (1) ? Distant, compact, astronomical radio source observed at frequency f – assume plane incident waves on the solar wind; and assume all source power is in a single point. ? Consider scattering by a single thin screen: assume solar wind varies in density by δN around mean density <N>. ? Mean refractive index (from Maxwell): n = 1 – fp2/2f2. ? Variation in refractive index: δn = δNe2/(8π2ε0 me f2); provided that f >> plasma frequency fp. assumed to a 1stapproximation that δN ∝ N2. ? The phase variation in a scattered wave in the solar wind can be ? The phase variation between different parts of scattered waves increases as N increases, and decreases as f increases. ? Next, consider the effect of a single thin screen in which the electron density varies sinusoidally (with wavenumber K)…
Basic IPS Theory (2) Taken from Daly, Ph.D. Thesis, 1999. ? An incident plane wave (wavenumber k) falling onto the x-plane. ? Figure shows the scintillation by a sinusoidal phase-changing screen: “weak scattering” (left) ∆φ << 1 radian, and “strong scattering” (right) ∆φ >> 1 radian; IPS becomes very complex in strong scattering and generally only when the weak-scattering regime applies do we use results from observations of IPS.
Basic IPS Theory (3) ? Providing the maximum variation in phase ∆φ << 1 radian (i.e. within the weak-scattering regime), the effect of the screen is to introduce a pair of additional waves (of complex amplitude i∆φ/2) propagating at angles of θ = ±sin-1(K/k) to the “average” wave. ? If K << k, that is, the scattering screen wavelength is much greater than the wavelength of the incident radio waves: θ = ±(K/k). ? If the waves come from a point source at infinite distance, then an observer looking through the screen from close to it would see the source flanked by a pair of weak images displaced by θ from it. ? The diffraction pattern (in intensity, or amplitude) across any plane “below” the screen “will result from the mutual interference of the three waves and will exhibit sinusoidal variations in amplitude and phase with wavenumber K” (Hewish, 1989). ? Thus, the diffraction pattern has same scale as a refracting screen.
Basic IPS Theory (4) ? Staying with the single sinusoidal scattering screen, then the screen has wavenumber wavelength: λ = 1/K. K, and ? If screen is moving across the ray path for the observation at speed V, then the diffraction pattern will also be moving across a receiver Solar Wind V “below” the screen at speed V. ? The receiver will record a time-varying intensity with peaks every 1/KV seconds. ? Observing the same source with two telescopes means the time difference at which peaks are observed can be used to determine V, which together with the time interval between peaks gives K.
Basic IPS Theory (5) ? However: the turbulence in the solar wind is not a simple, single, sinusoidal screen, and real radio sources are compact but rarely perfectly point-like. ? So, the next step is to look at a compact radio source as viewed through a screen which has random phase changes. ? For weak scattering, the spatial scale of fluctuations cast by the phase-changing screen is the same as the spatial scale of the irregularities in the screen. ? Phase change along path L through screen: θ = 2π/λ ∫Ln dz; then the thin diffraction screen imparts random phase variations to the emerging wavefront (amplitude initially unchanged). ? As the wavefront continues in the z direction, the interference leads to the formation of a spatial intensity pattern.
Basic IPS Theory (6) ? At low frequencies and in the weak-scattering limit, the intensity reduces to spectrum the phase spectrum multiplied by the Fresnel filter. ? The physical conditions for this case are that the scattered waves interfere only with the unperturbed (bulk-refracted) incident wave themselves. and not with
Basic IPS Theory (7) ? At low frequencies and in the weak scattering limit: the intensity spectrum P∆I(K)= 4sin2(K2z/2k)Pφ(K), where Pφ(K)is the phase spectrum which follows the spectrum of refractive index. ? If the scattering screen, and thus the resulting intensity pattern itself, drifts across the ray path, then it will be observed as a time series in signal power (i.e. as amplitude variations). ? The solar wind is a three-dimensional (3-D) structure, so radio waves from a distant compact radio source have passed through many scattering screens on their way from the source to where they are detected at the receiver. ? The assumption that the scattered waves interfere only with the unperturbed (bulk-refracted) incident wave and not with themselves is known as the Born approximation.
Basic IPS Theory (8) ? In weak scattering along with the Born approximation, the results of scattering from each individual screen can be treated independently of each other. ? Thus, the intensity pattern observed at the receiver is the linear sum of the intensity patterns generated by all scattering events along the ray path; this does NOT hold true for strong scattering. ? The ray path passes through extended regions of the solar atmosphere (solar wind structure) and properties of solar wind may (and indeed will) differ greatly. ? The ray path is nearest to the Sun near its “centre” and furthest away at the ends. ? The near-Earth regions of the ray path may not show fully- developed intensity variations.
Basic IPS Theory (9) ? This leads to the observed intensity spectra; correlation functions are weighted sums of spectra/correlation cast by each of the scattering events. functions ? We can (in principle) take account of weighting and reconstruct properties of each scattering screen from the observed intensity spectra and/or correlation functions.
Basic IPS Theory (10) ? The degree of variation of intensity seen in a time series generally scintillation index, m, as: m2= <(I(t) – <I>)>/<I>2, where I(t) is the source intensity observed at time t, expressed as and <I> is the average source intensity over the observation. ? The value of the scintillation index depends principally upon: the distance from the Sun (refractive index and thus phase variation); the source structure (at 928 MHz, sources wider than ~0.5” do not scintillate); the observing frequency; and the solar-wind density.
Basic IPS Theory (11)
Basic IPS Theory (12) The time lag for the peak of the cross- correlation function gives information on the solar- wind outflow speed and the ½ power time lags (skew of function) provide information on the spread in velocity(s). Spectral power contains information on irregularity scale-size distribution (and thus on solar wind density). The height of the cross- correlation function peak provides information on de-correlation, e.g. by Alfvén waves. The width of the auto- correlation function depends on the spatial scale of irregularities and solar wind speed.
Basic IPS Theory (13) IPS Power Spectrum. ; Single-frequency IPS by example. (Taken from: Bisi, Ph.D. Thesis, 2006.) Dual-frequency IPS. The Fresnel filter acts as a high-pass filter attenuating wave- numbers below the Fresnel spatial frequency, qf. For observing frequencies of 1,420 MHz (∼21 cm)/928 MHz (~32 cm)/500 MHz (~60 cm), the maximum scale-size of irregularities at a “thin screen” of scattering at 1AU distance from the Earth is ∼177 km/~219 km/~300 km, respectively. ?
Some Recent IPS Results and LOFAR Technical Advances.
IPS with LOFAR: The First CME Detection - R.A. Fallows, A. Asgekar, M.M. Bisi, A.R. Breen, S. ter-Veen, and on behalf of the LOFAR Collaboration, “The Dynamic Spectrum of Interplanetary Scintillation: First Solar Wind Observations on LOFAR”, Solar Physics “Observations and Modelling of the Inner Heliosphere” Topical Issue (Guest Editors M.M. Bisi, R.A. Harrison, and N. Lugaz), 285 (1-2), 127-139, 2013. - Bisi, M.M., S.A. Hardwick, R.A. Fallows, J.A. Davies, R.A. Harrison, E.A. Jensen, H. Morgan, C.-C. Wu, A. Asgekar, M. Xiong, E. Carley, G. Mann, P.T. Gallagher, A. Kerdraon, A.A. Konovalenko, A. MacKinnon, J. Magdalenić, H.O. Rucker, B. Thide, C. Vocks, et al., “The First Coronal Mass Ejection Observed with the LOw Frequency ARray (LOFAR)”, Submitted to The Astrophysical Journal Supplementary Series, June/July 2013 (and references therein).
The First CME with LOFAR… ? Observations of J1256-057 (3C279) detecting a CME with LOFAR on 17 November 2011 and (briefly) its comparison so far with other remote-sensing observations and modelling. Fully-consistent Results!
Preliminary LOFAR and EISCAT IPS Comparison Collaborative work with Dr. Richard A. Fallows, ASTRON, The Netherlands. An exact velocity match!!!
LOFAR Full Core IPS Dynamic Spectrum Courtesy of Dr. Richard A. Fallows, ASTRON. ? Two separate successive observations of IPS with radio source 3C48; strong scatter seen at progressively-lower frequencies.
KAIRA Ionospheric Scintillation More on ionospheric scintillation in the next Lecture by Norbert Jakowski. ? At low frequencies, the refractive effects starting to dominate at times. Courtesy of Derek MacKay/Richard Fallows, and KAIRA.
A Brief Introduction to IPS Three-Dimensional (3-D) Tomographic Reconstructions and Examples of their Usage. This 3-D tomographic technique was developed by Dr. Bernard V. Jackson et al., at the University of California, San Diego (UCSD), and is also available at NASA’s CCMC. For further details see Jackson et al., “Inclusion of Real-Time in-situ Measurements into the UCSD Time-Dependent Tomography and Its Use as a Forecast Algorithm”, Solar Physics, 285 (1-2), pp.151-165, doi:10.1007/s11207-012-0102-x, 2013, and references therein.