1 Title of proposed science use case
PROJECT DETAILS
Title
Low-Frequency Detection and Characterisation of Exoplanets
Principal Investigator
P. Zarka, J. Lazio, G. Hallinan et al.
Co-Authors
Exoplanet Team
Time Request
600 to 3000 hrs / Nbeams
>1000 exoplanets known, out of which up to 20% are interesting
targets based on theory (Zarka, Nichols…), previous surveys, and
observations at other wavelengths (e.g. stellar magnetic fields in
optics)
~400 stars and white and brown dwarfs within 10 pc, out of which
~200 known/observable, and ~2500 within 30 pc. We want to conduct
a significantly large imaging survey, observing a large fraction (200 to
500) of these targets ≥3 times x 1-2 hour each (with the 3
observations separated by a few hours or days to account for phase
effects or variability), i.e. up to 3000 hours. Multi-beam imaging – if
available - would reduce the effective number of observation hours.
FACILITY
Preconditions
Imaging (I & V) is more efficient for reaching a very high dynamic
range via calibration, while providing automatically many Off beams.
For candidates with potential radio emission identified by imaging,
follow-up observations could be made in coherent tied array beam
(TAB) mode with the entire array (filled core + as many surrounding
stations as possible) in order to observe flares and emission variations
at the orbital period of the exoplanet, that will be the definitive
evidence of exoplanet detection.
Coherent TAB observations requires at least 2 simultaneous beams
(On/Off). Multi-beam imaging with the entire array – if available would bring more efficiency. We need maximum sensitivity in each
beam, so that they should be formed with the whole array. But we can
trade bandwidth for beams number, forming 2 beams of ~125-150
MHz each, or more generally N beams of 250 or 300/N MHz each.
This option is standard in LOFAR software, requires no extra
correlator capacity, and should be easy to implement for SKA.
Alternately, follow-up observations of targets showing up in the
imaging survey can be made in imaging mode too.
x
SKA1-LOW
SKA1-MID
SKA1-SURVEY
RECEIVER(S) REQUIRED
x
SKA1-LOW
SKA1-MID Band 1
SKA1-MID Band 2
SKA1-MID Band 3
SKA1-MID Band 4
SKA1-MID Band 5
SKA1-SURVEY Band 1
SKA1-SURVEY Band 2
SKA1-SURVEY Band 3
Time (hrs)
600 to 3000 hrs / Nbeams
OPERATIONAL MODE
(as defined in Concept-of-Operations)
Normal
x
Fixed schedule (give cadence)
x
Time-critical override
Custom Experiment
Commensal
Collaborative & Coordinated
Sub-arrays required
Details
Not crucial as long as several
observations of the same target are
performed at various orbital phases, or
various times for a star that is not
known to host a planet
Systematically check the position of any
detected transient in any of the
observed fields versus the exoplanet
catalog
and
trigger
focussed
observations if a transient is detected in
coincidence
with
an
exoplanet’s
position.
Multi-beaming – if available - would
facilitate commensal observations.
No. Maximum sensitivity is crucial.
But bandwidth may be trade for beams
number (see above).
COMMENTS ON OBSERVING STRATEGY
200 to 500 targets to be observed (candidates from known exoplanets, selected magnetic stars,
stars closer than 10 pc, a fraction of stars closer than 30 pc), ~3 times x 1-2 hour each, separated
by a few hours or days (to track variations due to orbital phase, magnetically induced bursts or
response to flares).
Imaging with integrations of ~1 sec over the whole SKA1-LOW bandwidth is required, with
spectral resolution of order 1 MHz (post-RFI mitigation). Data may be integrated in broader bands
for higher sensitivity, but availability of a fair spectral resolution allows detailed studies of the
target when emission is detected.
Targets showing radio emission may be followed by longer duration beamformed observations
(with at least a beam on target and one off-beam), with similar resolutions. Alternatively, follow-up
observations could also be in imaging mode, but for searching for time variations in polar V on a
single target, beamformed observations may be more tractable.
Polarization (V mainly) is important for lowering confusion noise and separating stellar from
planetary emissions.
Dense exoplanet fields (Corot, Kepler) are good targets for multi-planet search of any orbital
period of a knwon planet in the field.
Multi-beaming imaging and/or TAB will increase efficiency. Nbeams can be traded with overall
observation bandwidth (e.g. 5 beams x 50-60 MHz – see above).
POLARISATION PRODUCTS REQUIRED : BEAMFORMER ( X ) or CORRELATOR (
x Stokes I
XX
YY
Stokes Q
XY
Stokes U
x Stokes V
YX
)
SCIENTIFIC DESCRIPTION (max 200 words)
Solar system magnetized planets produce extremely bright, highly circularly polarized, coherent
radio emission at low frequencies, attributed to electron cyclotron maser emission. Detection and
characterization of this emission:
‐ allows measurement of the planetary magnetic field strength, tilt & offset (radio emission is
produced at the electron cyclotron frequency)
‐ permits measurement of the true rotation rate of planet and the magnetosphere-satellites
coupling
- provides indirect insight into the dynamics and composition of the planetary interior.
Detection of similar emission from an exoplanet represents the most promising method for
directly measuring their magnetic fields, and constrain orbit inclination, star-planet plasma
interaction, possible existence of satellites, and habitability (magnetosphere limits atmospheric
erosion & destruction of O3 by cosmic rays). Hot Jupiters are a particularly promising population
for first detection.
Radio emission from an exoplanet is narrowly beamed, therefore an observing campaign needs
to monitor a large number of planets to account for geometrical selection effects , and sample the
rotation period to detect the rotationally modulated emission.
Tidally locked hot Jupiters can have rotation periods of a few days, requiring observations on
minute timescales (to detect bursts) over hours, repeated hours to days apart. Observations of
brown dwarfs indicated that this kind of radio emission is detectable.
Jupiter-like emission will be detectable with SKA up to a few 10s of parsecs distance.
‘TARGETS’ OF OBSERVATIONS
Type of observation
x Individual pointings per object
(what defines a ‘target’)
x Individual fields-of-view with multiple objects
x Maps through multiple fields of view
x Non-imaging pointings
Number of targets
200 to 500
Positions of targets
Rapidly changing sky position?
(e.g. comet, planet)
Time Critical?
Integration time per target
(hrs,mins,secs as appropriate)
Average peak flux density
(Jy per beam per Hz)
Range of peak flux densities
(Jy per beam per Hz)
Expected polarised flux density
(expressed as % of total)
All Sky (isotropic).
YES [details: ………………………………………..]
x NO
YES [details: ………………………………………..]
x NO
1 s integrations, but 1 hour on target per observation
10s to 100s of micro-Jy.
Highly variable
100
OBSERVATIONAL SETUP : BEAMFORMER ( ) or CORRELATOR ( )
Full LOW
range, except trade-off
Central Frequencies (MHz)
bandwidth versus Nbeams
(including redshift, observatory correction)
250-300 MHz (/ Nbeams)
Total Bandwidth (MHz)
1 MHz (post RFI mitigation)
Spectral resolution (kHz)
Temporal resolution (‘dump’ time in s or 1 sec is adapted for imaging and RFI
mitigation baseline by baseline; in tied
‘standard’)
array beam mode, need of higher time and
frequency resolutions (~10 msec x a few
kHz)
NON-IMAGING SPECIFIC CONSIDERATIONS
Absolute flux scale calibration
IMAGING CONSIDERATIONS (continuum – single channel of full bandwidth)
Required angular resolution (arcsec)
(single value or range)
Required image size (arcsec)
(single value or range)
Required rms (Jy per beam)
(if polarisation products required define for each)
Dynamic range within image
(if polarisation products required define for each)
Absolute flux scale calibration
1-3%
5%
10%
20-50%
x n/a
IMAGING CONSIDERATIONS (spectral – multiple channels of narrow bandwidth)
Point sources at resolution of ~20” at 60
Required angular resolution (arcsec)
MHz.
(single value or range)
~5 degrees FoV
Required image size (arcsec)
(single value or range)
10 channels between 60 and 260 MHz
Number of image channels
20 MHz
Channel width (kHz)
RFI mitigation must be performed at finer
spectral scale.
In case of detection, a dynamic spectrum
must be obtained at a resolution much
better than 20 MHz (~1 MHz)
At intermediate frequencies (100 MHz), a
Required rms (Jy per beam per channel)
(if polarisation products required define for single 1 hour observation should provide an
rms of 10 microJy per beam.
each)
Dynamic range within image per channel
(if polarisation products required define for
each)
Absolute flux scale calibration
1-3%
x 5%
10%
20-50%
n/a
DATA ANALYSIS
Procedures required
Processing considerations
(e.g. flag high wind speed data,
reprocessing required?)
Data products
Initially bin data to 2-4 kHz, sum spectra coherently
from all stations (keep orthogonal polarizations
separate). In depth RFI mitigation. Time-frequency
integration. Calibration & Imaging.
Dedispersion – measured DMs are < 2 pc cm-3, so try 510 values below this number. Integrate in time (~ 5
minutes) and frequency using MFS (~ 20 MHz).
The slow transient pipeline may also be used for
beamformed observations.
See above
Images, time series, full band dynamic spectra, Stokes I
& V.
Description of pipeline
Quality assessment plan & cadence
ISSUES TO BE DETERMINED/RESOLVED
(Here you should include any additional information that needs to be resolved before this
science can be carried out)
REFERENCES
• Poster “All the Sky, All the Time” by Gregg Hallinan presented at
Planetary Magnetic Fields: Planetary Interiors and Habitability, 2013
• Presentation “Exoplanet searches in Radio: theory & searches with UTR2, GMRT and LOFAR”
by P. Zarka at the SKA-Col workshop, Jodrell Bank, Nov. 2013.
[Willes & Wu, 2004, 2005] Theoretical approach : predicting radio spectra from unipolar inductor + CMI
[Zarka et al., 2001, 2004 ; Zarka 2007, 2010, 2011] Heuristic approach : predicting radio fluxes from
scaling law
[Hess & Zarka, 2011] Theoretical approach : predicting radio dynamic spectra from CMI modelling for
various SPI scenarii
[Nichols 2011, 2012] Radio flux predictions for normal Jupiters around highly XUV luminous stars  large
radio fluxes requires rapid Xo rotation (1-3h).
[Lazio 2010, Hallinan 2013] Past negative searches.
References
Willes, A.J., Wu, K., 2004. Electron-cyclotron maser emission from white dwarf pairs and white dwarf
planetary systems. MNRAS 348, 285–296.
Willes, A.J., Wu, K., 2005. Radio emissions from terrestrial planets around white dwarfs. Astron.
Astrophys. 432, 1091–1100.Thèse R. Laine
Hallinan, G., S. K. Sirothia, A. Antonova, C. H. Ishwara-Chandra, S. Bourke, J. G. Doyle, J. Hartman, and A.
Golden, Looking for a pulse: a search for rotationally modulated radio emission from the hot
jupiter, τ Bootis b, Astrophys. J., 762, 34, 2013.
Hess, S. L. G., and P. Zarka, Modeling the radio signature of the orbital parameters, rotation, and magnetic
field of exoplanets, Astron. Astrophys., 531, A29, 2011.
Lazio, T. J. W., P. D. Shankland, W. M. Farrell, and D. L. Blank, Radio observations of HD 80606 near
planetary periastron, Astron. J., 140, 1929-1933, 2010.
Nichols, J. D., Magnetosphere–ionosphere coupling at Jupiter-like exoplanets with internal plasma
sources: implications for detectability of auroral radio emissions, Mon. Not. R. Astron. Soc., 414,
2125–2138, 2011.
Nichols, J. D., Candidates for detecting exoplanetary radio emissions generated by magnetosphere–
ionosphere coupling, Mon. Not. R. Astron. Soc., 427, L75–L79, 2012.
Zarka, P., R. A. Treumann, B. P. Ryabov, and V. B. Ryabov, Magnetically-driven planetary radio emissions
and applications to extrasolar planets, Astrophys. Space Sci., 277, 293-300, 2001.
Zarka, P., B. Cecconi, and W. S. Kurth, Jupiter's low frequency radio spectrum from Cassini/RPWS absolute
flux density measurements, J. Geophys. Res., 109, A09S15, 2004.
Zarka, P., Plasma interactions of exoplanets with their parent star and associated radio emissions, Planet.
Space Sci., 55, 598-617, 2007.
Zarka, P., Radioastronomy and the study of exoplanets, « Pathways Towards Habitable Planets », ASP
Conference Series, Vol. 430, V. Coudé du Foresto, D. M. Gelino and I. Ribas, eds., 175-180, 2010.
Zarka, P., The search for exoplanetary radio emissions, in "Planetary Radio Emissions VII", H. O. Rucker et
al. eds., Austrian Acad. Sci. Press, Vienna, pp. 287-302, 2011.