Will amateur radio astronomers be the first to directly detect extrasolar planets?
Robert Clark
wobbling seen in some stars. Their actual light still has not been
detected or distinguished from that of their parent stars.
The long wavelength radio bursts that emanate from Jupiter have led
to suggestions that extrasolar planets might be detected by searching
for such bursts in the vicinity of stars:
Opening a New Window on the Universe:
High Resolution, Long Wavelength Radio Astronomy,
2.5.2 Extrasolar Planets,
by Joseph Lazio
http://rsd-www.nrl.navy.mil/7213/lazio/decade_web/node34.html
However, sensitive searches have so far failed to detect them. This
is undoubtedly due to distance attenuation for such planets
light-years away. The distance to Jupiter ranges up to 9 x 10^8 km. A
star 10 light-years away is at 9 x 10^13 km, a factor of 10^5 larger
than the Jupiter distance.
The Jovian radio bursts have been detected by amateurs with simple
dipole antennas:
Radio-Jupiter for Amateur Observers, By Jim Sky
http://web2.thesphere.com/SAS/bulletin/Sas44/page1.html#Jupiter
PROJECT P5-2. JUPITER-IO MAGNETOSPHERE RADIO NOISE
http://www.elmag5.com/jupiter-io.htm
With its Radio JOVE project NASA also distributes low-cost dipole
kits to schools:
How To Hear Radio Signals From Jupiter
http://www.spacetoday.org/SolSys/Jupiter/JupiterRadio.html
Radio JOVE
http://spacescience.com/headlines/y2000/ast22may_1.htm?list
The proposal is for amateur radio astronomers to set up arrays of
such low cost dipole antennas world-wide. The T-shaped dipoles have
the advantage of steerability, but the vertical dipoles have the
advantage of simplicity and low cost for setting up large arrays. An
example of a steerable dipole array is the one that first discovered
the Jovian decametric emissions:
The Discovery of Jupiter's Radio Emissions
How a chance discovery opened up the field of Jovian radio studies
http://radiojove.gsfc.nasa.gov/library/sci_briefs/discovery.html
The signals from the various arrays would be combined digitally to
form a world-wide radio telescope. The large-wavelengths being
detected simplify the task of combining the signals
interferometrically. GPS transmitters are now available that can give
locations to within inches:
NASA satellite technology goes down on the farm
http://www.spaceflightnow.com/news/n0105/11farm/
The clocks in such transmitters also provide timing at better than
nanosecond resolution.
The signal strength for Jupiter at 10 light-years would decrease as
the square of the distance, so would be smaller by a factor of 10^10.
However, the extrasolar Jovian planets detected so far have been close
in to their primaries and are expected to produce stronger radio
emissions than Jupiter, perhaps, 100 to 1000 times more intense. Using
the optimistic estimate of 1000 times greater intensity would require
100,000 separate arrays with 100 dipoles or 10,000 arrays with 1,000
dipoles to detect such emissions.
Bob Clark
William F. Hagen
some salesmen will do anything to keep a client.
Mike Williams
>The existence of extrasolar planets has been inferred from the
>wobbling seen in some stars. Their actual light still has not been
>detected or distinguished from that of their parent stars.
Four Brits may have done so:-
CAMERON A., HORNE K., PENNY A. & JAMES D., 1999
Probable detection of starlight reflected from the giant exoplanet
orbiting tau Bootis. Nature, 402, 751
<http://www.nature.com/server-java/Propub/nature/402751A0.frameset?context=toc>
Also there have been some definite observations of transits, when the
light from the star has been observed to be dimmed by the passing of a
planet across its face, such as
CHARBONNEAU D., BROWN T., LATHAM D., MAYOR M. & MAZEH T., 1999b
Detection of Planetary Transits Across a Sun-like Star
ApJ., 529, 45
--
Mike Williams
Gentleman of Leisure
Alan Anderson
> The existence of extrasolar planets has been inferred from the
> wobbling seen in some stars. Their actual light still has not been
> detected or distinguished from that of their parent stars.
Um...last I heard, several such planets indeed had been imaged.
webmaster
"Alan Anderson" <aran...@netusa1.net> wrote in message
news:aranders-230...@rash4-168.kokomo.in.hypervine.net...
No, not yet, although there have been several Brown Dwarfs that have been
imaged. Astronomers are still debating where the line exists between the
two.
Nathan www.livefromspace.tv
T. Joseph W. Lazio
RC> The existence of extrasolar planets has been inferred from the
RC> wobbling seen in some stars. Their actual light still has not been
RC> detected or distinguished from that of their parent stars. The
RC> long wavelength radio bursts that emanate from Jupiter have led to
RC> suggestions that extrasolar planets might be detected by searching
RC> for such bursts in the vicinity of stars:
RC> Opening a New Window on the Universe: High Resolution, Long
RC> Wavelength Radio Astronomy, 2.5.2 Extrasolar Planets, [...]
RC> http://rsd-www.nrl.navy.mil/7213/lazio/decade_web/node34.html
RC> However, sensitive searches have so far failed to detect
RC> them. This is undoubtedly due to distance attenuation for such
RC> planets light-years away. The distance to Jupiter ranges up to 9 x
RC> 10^8 km. A star 10 light-years away is at 9 x 10^13 km, a factor
RC> of 10^5 larger than the Jupiter distance.
The key is to go to low frequencies (low by radio astronomical
standards). Jupiter's emission cuts off quite strongly around 40
MHz. Thus, most professional radio telescopes, which are optimized to
work around 1000 MHz have little chance of detecting any radio
emission from these exoplanets.
As you say, there has been no detection of radio emission from an
exoplanet. I know that Bastian and collaborators looked, using the
VLA, toward many stars. However, most of their observations were at
330 MHz and above; again one might worry that at these high
frequencies the exoplanets produce no emission anyway.
I've got a project going to look at tau Boo. At least according to
one model, there's a slim chance that it would produce detectable
emission at 74 MHz, a frequency at which the VLA does operate. Stay
tuned! :)
--
Lt. Lazio, HTML police | e-mail: jla...@patriot.net
No means no, stop rape. | http://patriot.net/%7Ejlazio/
sci.astro FAQ at http://sciastro.astronomy.net/sci.astro.html
Chosp
"Alan Anderson" <aran...@netusa1.net> wrote in message
news:aranders-230...@rash4-168.kokomo.in.hypervine.net...
Where did you hear this?
Jan Panteltje
(Robert Clark) wrote in <5f181508.01052...@posting.google.com>:
I think it is a cool idea.
Not sure about where to buy an atomic clock..
Timing may really become a problem.
(Did not read all the links yet, tonight perhaps).
Regards
Jan
Alan Anderson
> > Um...last I heard, several such planets indeed had been imaged.
>
> No, not yet, although there have been several Brown Dwarfs that have been
> imaged. Astronomers are still debating where the line exists between the
> two.
Then what was all the hubbub a couple of years ago with the Hubble and the
first direct view of a planet around another star? I remember it
distinctly, with it described as a planet several times larger than
Jupiter at about a thousand AU from a binary star almost hidden by a dust
cloud. Was it just hype?
Mike Williams
>Then what was all the hubbub a couple of years ago with the Hubble and the
>first direct view of a planet around another star? I remember it
>distinctly, with it described as a planet several times larger than
>Jupiter at about a thousand AU from a binary star almost hidden by a dust
>cloud. Was it just hype?
There was a famous case called TMR-1C of something that looked like it
might possibly be a planet in a Hubble image. The only reason for
thinking that it was a planet rather than a background star was that
there was a dust filament that sort of spiralled out from the star and
ended exactly at the "planet", and the theory was that the dust and
planet had been flung out from the star. It turns out that the object is
far too bright to be a planet illuminated only by reflected starlight.
The best guess is that it's a more remote star that just happens to be
in the same line of sight as the end of the filament.
A websearch for TMR-1C still throws up a lot of sites announcing the
detection of a possible image of a planet, and remarkably few that
mention that it turned out not to be a planet.
George Emsden
orbiting round another "Sun" is going to help us find ETI?
As techniques improve, I am sure that more planets will be discovered but
this does not get us closer to discovering ET civilisations where the SETI
project is basically trying to process buckets of radio signals to sort out
the natural from the "man-made" for want of a better word.
--
George Emsden
"Robert Clark" <rgc...@my-deja.com> wrote in message
news:5f181508.01052...@posting.google.com...
T. Joseph W. Lazio
GE> "Robert Clark" <rgc...@my-deja.com> wrote in message
GE> news:5f181508.01052...@posting.google.com...
>> The existence of extrasolar planets has been inferred from the
>> wobbling seen in some stars. Their actual light still has not been
>> detected or distinguished from that of their parent stars. The
>> long wavelength radio bursts that emanate from Jupiter have led to
>> suggestions that extrasolar planets might be detected by searching
>> for such bursts in the vicinity of stars:
>>
>> Opening a New Window on the Universe: High Resolution, Long
>> Wavelength Radio Astronomy, 2.5.2 Extrasolar Planets, [...]
>> http://rsd-www.nrl.navy.mil/7213/lazio/decade_web/node34.html
GE> Could someone tell me how the existence of a large planet like
GE> say, Jupiter orbiting round another "Sun" is going to help us find
GE> ETI?
Nobody has claimed that it will, at least not directly. There is the
indirect aspect that finding other solar systems does improve our
confidence in planets being a natural by-product of star formation.
I think there's also some excitement in being able to study planets in
another solar system. Current detection techniques allow us to say
that there is a planet and to constrain its mass but not much else.
Being able to detect and study the natural radio emission from planets
in another solar system would allow us to begin studying the planets
as bodies themselves.
GE> As techniques improve, I am sure that more planets will be
GE> discovered but this does not get us closer to discovering ET
GE> civilisations where the SETI project is basically trying to
GE> process buckets of radio signals to sort out the natural from the
GE> "man-made" for want of a better word.
Another indirect aspect of being able to detect the natural radio
emission from exoplanets is the likelihood that we will have to
develop more clever ways of filtering out human-generated
interference. These improved techniques might also have considerable
impact on the "sorting" or filtering that radio SETI programs have to
do.
Alfred A. Aburto Jr.
This is a great idea! Have you talked to the "Radio Jove" group ... a world
wide group monitoring Jupiter emissions. Also there are people in SARA
(Society of Amateur Radio Astronomers) monitoring Jupiter. They'd be excited
about looking for the 20MHz like emissions from other giant planets! I guess
the only problem to do coherent or incoherent summing from all those amatuer
radio telescopes (antennas :) would be to have accurate clocks at each site.
One question though. Would the emissions always be in the 20MHz region?
Doesn't the emission at 20MHz depend upon the planet having a nearby moon
(Io in Jupiters case)?
Al
> T. Joseph W. Lazio <jla...@patriot.net> wrote in message
news:m2lmnmg...@patriot.net...
Robert Clark
Cameron et.al. seemed to have retracted their claim; at least they
now acknowledge they can't confirm it:
TAU BOO b – NOT SO BRIGHT, BUT JUST AS HEAVY
Introduction
"Here we present new optical data secured in spring 2000. These show
the earlier reflection candidate to have been spurious, but confirm
that the star co-rotates with the planet’s orbit."
http://star-www.st-and.ac.uk/~acc4/tboopages/ManPoster/tsld002.htm
If the sensitivity of this method can be improved to make confirmed
detections, then I think most astronomers would consider this to be an
actual detection of light from an exoplanet since you could make
spectroscopic determinations of the content of its atmosphere.
Bob Clark
Joseph Lazio
AAA> This is a great idea! Have you talked to the "Radio Jove" group
AAA> ... a world wide group monitoring Jupiter emissions. Also there
AAA> are people in SARA (Society of Amateur Radio Astronomers)
AAA> monitoring Jupiter. They'd be excited about looking for the 20MHz
AAA> like emissions from other giant planets! I guess the only problem
AAA> to do coherent or incoherent summing from all those amatuer radio
AAA> telescopes (antennas :) would be to have accurate clocks at each
AAA> site.
Yes. I think to have any hope of detecting these emissions, one needs
an array. At these frequencies, "confusion" becomes a severe
problem. ("Confusion" occurs when the telescope does not have
adequate resolution to separate sources. The result is that one's
sensitivity can be much worse than one might otherwise expect.) Only
by considering a distributed set of antennas can one obtain enough
resolution to defeat confusion. In this case, one probably wants the
antennas distributed over a 50--100 km area.
As you note, there are considerable infrastructure problems that would
need to be addressed. One either needs to provide a common time or
frequency standard for all of the antennas or ensure that all antennas
have a highly accurate clock. There's also the issue of how to
communicate the signals from the individual antennas to a central
processing facility so the antennas can be (virtually) linked together
to form the array. Finally, there are some severe computation
problems. The "heart" of any radio telescope array is a central,
high-speed, special-purpose computer known as a correlator. There is
no correlator in existence today that could handle the large number of
antennas whose signals would need to be processed. >sigh< As you
might guess by now, little is easy.
AAA> One question though. Would the emissions always be in the 20MHz
AAA> region? Doesn't the emission at 20MHz depend upon the planet
AAA> having a nearby moon (Io in Jupiters case)?
I don't think so. Io certainly contributes to some of Jupiter's radio
emission. My understanding is that some fraction of Jupiter's
emission is also driven by the solar wind-magnetosphere interaction,
though. Thus, even in the absence of Io, Jupiter would still produce
some radio emission.
The range of frequencies in these models is determined by both the
solar wind and the magnetosphere. As many of the exoplanets are
comparable to or larger than Jupiter, it is assumed that their
magnetospheres are as large as Jupiter's. Furthermore, they are
closer to their primary stars, increasing the effective stellar wind
pressure on their magnetospheres.
The result is that radio emission at frequencies higher than 40 MHz
certainly seems plausible. Whether it exists or not, is a different
issue! :)
friedt jean-michel
> need to be addressed. One either needs to provide a common time or
> frequency standard for all of the antennas or ensure that all antennas
> have a highly accurate clock. There's also the issue of how to
I do not know what resolution is required on synchronisation of the
various frequency references used on the various radiotelescopes, but
following the work developed at the Observatoire de Besancon (France)
for the Auger particle detector, I had a look at some of the GPS
based standards and was quite impressed by
http://www.rt66.com/~shera/index_fs.htm
Synchronizing a PLL on the 1pps signal of the GPS does not seem
awfully difficult, though as I have never tried it there might
be unexpected problems. Calibrating the resulting frequency standard
is another complex problem (where to access a Cs or Rb reference clock ?
Any available on the surplus market ?)
I have been told by the people of the Observatoire de Besancon that
getting the 1pps synchronization is not too difficult and can be
done using commercial receiver, but synchronizing on the carrier is
much more difficult and expensive.
> communicate the signals from the individual antennas to a central
> processing facility so the antennas can be (virtually) linked together
I suppose post-processing and sharing via the internet can solve
that part (once you have your precise shared frequency standard), or
is real-time processing required ?
Paul Bramscher
> The result is that radio emission at frequencies higher than 40 MHz
> certainly seems plausible. Whether it exists or not, is a different
> issue! :)
In an ideal world, we'd have discovered thousands of earth-sized planets,
and aimed a space-based Arecibo-sized telescope directly at them, scanning
thousands of frequencies.... I still think our essential problems with
S@H are (1) no control of the aim of Arecibo and (2) too limited a
frequency scan. We're literally shooting in the dark at a moving target.
In reverse, if intelligent life (neither too far advanced, nor too far
behind) was common and using radio, SETI would be another method of
detecting non-jovian planets. Although it's possible that intelligent
radio emissions could be coming from a deep-space transmitter or spaceship
enroute I imagine.
Robert Clark
timing requirements for interferometry at decametric wavelengths:
"The timing accuracy question really
has two parts - I should have been more clear in my
previous message. To coherently combine two signals
from distant antennas, you need to digitize and time
tag the signals with local oscillators that are accurate
enough so the phase difference between them stays within
a radian or so for a reasonable length of time. This is
the coherence time, and is approximately equal to the
observing frequency in Hz divided by the fractional
frequency stability of the oscillators. Good quality
crystal oscillators have stabilities of around 10E-8,
so the coherence time for observations at 10 MHz would
be about 10 seconds. (I'm ignoring factors of sqrt(2)
and the fact that the ionosphere may often impose a
shorter coherence time at this low a frequency.)
Once you have time tagged the signal samples, you
need to line up the bit streams so the bits corresponding
to the same wave front are being combined. This is
where the bandwidth comes in: the minimum sampling
time is the Nyquist rate, which is twice the bandwidth
of the signal you are sampling. Thus, the time between
samples will be 1/(2 * bandwidth) seconds. So for a
wide bandwidth like 10 MHz you need timing accuracy of
0.05 microseconds (I had 0.1 microsec in my previous
message by mistake). For a narrower bandwidth like
100 kHz you only need timing to 5 microseconds accuracy.
In either case you need local oscillators that are
stable enough to give you a useful coherence time at
your observing frequency. I hope this is helpful.
Cheers,
Dayton"
I was also informed that the GPS transmitter I mentioned, from Navcom
Technology, has timing accuracy at the sub-nanosecond level, which
should be sufficient for decametric wavelengths. Typically handheld
GPS transmitters cost in the $100 to $300 range, though I imagine the
Navcom transmitter would be a bit pricier.
Dr. Jones also brought up the computational problem of combining that
many signals digitally. Perhaps the method of distributed computing
could work, the method used with Seti@home. An article describing an
Intel sponsored distributed computing project says the planned 6
million PC's would result in computing power 10 times greater than the
largest supercomputer:
Use your PC to fight cancer, hunt aliens and more
"The Intel program, launched in April, hopes to recruit 6 million
people to download its client software. Their PCs will provide 50
terraflops of collective processing power (1 terraflop equals 1
trillion floating-point computations per second). "That's ten times
bigger than the world's largest supercomputer ... for less than 1
percent the cost," says Pat Gelsinger, Intel's chief technical
officer."
http://www.cnn.com/2001/TECH/ptech/05/21/pc.uses.idg/index.html
Dr. Phillipe Zarka also directed me to a site with links to research
on long wavelength astronomy:
Prospective en Radioastronomie Basse Fréquence
à l'Observatoire de Paris
http://despa.obspm.fr/~zarka/LF-Radioastronomy/
In particular two articles co-authored by Dr. Zarka discuss the
detection of radio emissions from extrasolar planets:
P. Zarka, R.A. Treumann, B.P. Ryabov & V.B. Ryabov:
Magnetically-driven planetary radio emissions and applications to
extrasolar planets, Les rencontres de líObservatoire "Physics of
space: growth points and problems", Astrophys. Space Sci., in press,
2001. (PS 104 Ko)
http://despa.obspm.fr/~zarka/LF-Radioastronomy/Obs2000.ps
P. Zarka, J. Queinnec, B.P. Ryabov, V.B. Ryabov, V.A. Shevchenko, A.V.
Arkhipov, H.O. Rucker, L. Denis, A. Gerbault, P. Dierich & C. Rosolen:
Ground-Based High Sensitivity Radio Astronomy at Decameter
Wavelengths, in "Planetary Radio Emissions IV" (Graz, 9/1996), H.O.
Rucker, S.J. Bauer & A. Lecacheux Eds., Austrian Acad. Sci. press,
Vienna, p. 101-127, 1997. (PDF 768 Ko)
http://despa.obspm.fr/~zarka/LF-Radioastronomy/Ground-based.Graz.pdf
In the latter paper in section 3.2, the authors calculate that
close-in Jovian planets might have decametric emissions up to 8x10^5
times greater than the Jupiter decametric emissions, though they state
that typical they might be about 1,000 times greater.
Bob Clark
rgc...@my-deja.com (Robert Clark) wrote in message news:<5f181508.01052...@posting.google.com>...
David Woolley
rgc...@my-deja.com (Robert Clark) wrote:
> JPL radio astronomer Dr. Dayton Jones responded to my question on the
> timing requirements for interferometry at decametric wavelengths:
> > and the fact that the ionosphere may often impose a
> > shorter coherence time at this low a frequency.)
Also, the ionospheric effect will be much different at
GPS frequencies, so using GPS as a time reference will
not give a good prediction at these frequencies.
> > In either case you need local oscillators that are
> > stable enough to give you a useful coherence time at
> > your observing frequency. I hope this is helpful.
Coherence length is a better measure than coherence time. You
need a coherence length that is maybe 65,000 km (10 times earth
radius). That's about quarter of a second in time, so about
0.4 ppm stability (maybe a bit better because there are two
parties) is enough at 10MHz.
> I was also informed that the GPS transmitter I mentioned, from Navcom
> Technology, has timing accuracy at the sub-nanosecond level, which
What is mentioned is a receiver, not a transmitter. It uses the normal
GPS transmitters, pluse information transmitted on the internet about
imperfections in the orbit data and clock frequencies of those
satellites, by reference to measurements made by accurately surveyed
earth stations.
> should be sufficient for decametric wavelengths. Typically handheld
> GPS transmitters cost in the $100 to $300 range, though I imagine the
Hand held GPS transmitters are electronics countermeasures devices and
unlikely to be sold to the general public!
THe specified time transfer accuracy for commercial GPS receivers is around
50ns, with larger short term excursions (the short term excursions are
important in this application, unless you have an atomic clock to smooth
them).
In reality, most hand helds are not designed for time transfer and will
give nothing like that precision because of software scheduling delays
in providing the time to the outside world. Some relatively cheap receivers
are designed for time transfer and will give an accurate one pulse per second
signal. I doubt that the signal characteristics are well enough defined
for sub-nano second reading - you would probably need more than 1pps as well
as very fast edges.
> Navcom transmitter would be a bit pricier.
It's not clear that the receivers in question are available to the general
public. The ones used for the performance data use the military second
channel. It's not clear whether they are talking of full military spec
receivers, using the full military precision, or some hack that recovers
some additional information off the L2 signal carrier without actually being
able to de-spread the spreading code. There are such receivers, but it wasn't
clear whether the performance data was for one of these, or for a proper
military receiver.
Even with a military receiver, you need one designed for time transfer
purposes, not for location purposes.
The very need for the L2 signal is an indication that ionopsheric variation
with frequency is an issue.
> Dr. Jones also brought up the computational problem of combining that
> many signals digitally. Perhaps the method of distributed computing
More of a problem is the volume of data. Anyone contributing would
need a commercial, possibly, an ISP level, internet connection as they
would be uploading large volumes of data continously. If you did
do this, and kept the processing in the family, the best approach would
allocate time ranges to participants and to directly transmit all data
for a particular time range to the person processing it (you will need
to buffer the data - even if you transmit real time, the receiver will
not be able to cope with everyone sending data to one place at once,
and scheduling delayed transmission, to even out the load, is probably
more efficient than running multiple slow connections.
In terms of commercial, centralised, distributed computing, the data
volumes would probably be too high for the project to be worth considering.
To make it worthwhile, you would have to process only a small part
of the available data.
Also, for a centralised model, you would have to demonstrate that
there really was enough processing done per unit time to offset the
distribution costs. It is possible that there is, as one needs to
not just add the signals, but add them for a large number of different
possible source directions. Because the array is not on a plane and
is rotating around the earth, this may not be possible using fast
algorithms.
The real problem, though, with all this is that you will never get the
number of receiving stations proposed for a minimum system. One only has
to look at the number of active Project Argus stations, with a similar
cost and using rather less in the way of non-standard amateur radio
equipment, to realise that there are not that many people prepared to
invest sufficient resources in such a project.
S@H gets large numbers of users because people don't perceive any marginal
cost (additional electricity costs are not obvious to most) and those
who do invest for S@H, for the most, do it for the competition element,
not for the science.
> Newsgroups: rec.radio.amateur.space,rec.radio.amateur.antenna,sci.astro,sci.astro.seti,sci.space.policy
Cross-posting r.r.a.space and r.r.a.antenna violates a principle that
you do not cross-post to siblings or parents and children, although not
reading this there, I'm not sure which should be dropped.