Search for Planet Barbarossa

[Joe Keller]

Object
New planet ?
Coordinates 2000.0
RA: 11h 26m 49s DEC: -09º04’21”
Telescope
Cassegrain Relay 400 mm
Focal length /Ratio F/Field
2614 mm/ 6.5 / 18’ x 12’
Site/Position
Tacande Observatory
Longitude : 17g 52m 03,55s W
Latitude: 28º 38’ 29.79” N
Height: 765 m.
CCD/Scale/Spectrograph *
ST8XE 0.71”/pixel
Integration details
1 x 10 minutes
Date(d/m/y)/Time (UTC)
25/03/2007 00:41:39
Moon phase %
52.5
Airmass
1.29
Calibration spectra *

Calibration source *

Software
Maxim DL/CCD
Operator/s
Joan Genebriera
Comments
Coordinates of the star GSC 5509:1104

* Spectroscopic only

If you email me, I'll try to email you the photo (sometimes that works).

[Joe Keller]

In this 0.7" resolution photo by Joan Genebriera (electronic, not a scan of film) [the asteroid] is north of the date "25" printed in the lower right corner. [The asteroid, seen as two points of light] is nearer the edge than the bottom. Together with the three stars to the NW, they make a line of 5 points of light.

Sincerely,
Joseph C. Keller, M. D.

[Joe Keller]

Dear Ms. Genebriera & Mr. Riley,

I compared all of Mr. Riley's photo to its counterpart, Ms. Genebriera's photo "Barbarossa_3" (i.e., the photo with the more northerly coordinates, on which I saw the candidate object). Also I compared them to the SERC DSS2 (Red filter) image in the Aladin archive. Both photos showed excellent correspondence to the archive image.

Other than the candidate object on Ms. Genebriera's photo, I saw nothing on either photo that did not match the DSS2 image (except for obvious very slight defects). In particular, I saw nothing retrograde thereof on Mr. Riley's photo.

I spoke lengthily yesterday with *********, to whom I had emailed Ms. Genebriera's photo. His opinion was that the candidate object was *not* a cosmic ray artifact.

So, it might have been an asteroid on Ms. Genebriera's photo. An asteroid would be out of the field of view of Mr. Riley's photo. A trans-Neptunian object, even as close as 30 AU from the sun (which would give 5 arcminutes motion in 3.2 days now, near opposition) would have been inside the field of view (centered on the candidate object).

Thank you both for your assistance. This initial negative result neither proves nor disproves the existence of a distant planet shepherding a point of the 5:2 Jupiter:Saturn resonance. The 1987 SERC image I discovered, is consistent with such a planet.

I'll forward to both of you, any important information I acquire in the future about this. Meanwhile, if either of you take more photos along the ecliptic in this area, I will give my full attention to their analysis.

Sincerely,
Joseph C. Keller, M. D.

[higginsdj]

Hi Joe,

Your enthusiasm is admirable (and there are very likely many large objects still out there lying undiscovered) but your observational assessments leave a lot to be desired. You really do need to learn the scientific method and the appropriate mechanisms for confirming and reporting things. Now the next time you report 'finding' something will likely reap more sceptasism that this occurance. The methods are there for a reason..... Without observational experience you might like to leave the interpretation and analysis of images to those with more experience in such work!

Cheers

[Joe Keller]

I found another disappearing dot, again on a scanned archival Red plate (absent from two other archival Red scanned plates, also one archival Blue and one archival Optical Infrared). Its Red magnitude is +17.8 by comparison with a nearby star with stable catalog magnitudes. Its coordinates are
RA 11 03 12.4 Decl -5 58 09
This is a POSS I Palomar plate from 1954.154. The position is quite consistent with the track, the period and the 1987 archival object (est. Red mag +17.3) discussed above.
Tonight's coordinates, based on great-circle extrapolation & corrected for Earth parallax are
RA 11 25 49.4 Decl -9 03 02.

[higginsdj]

HI Joe,

May I ask by what means you are measuring the images and how you are determining correlation or consistancy with the track of your proposed object? What level of error do you have in your measurements? To extrapolate a position implies you have determined a rough orbit - do you feel inclined to share the orbital parameters?

Is ther extrapolation based on just 2 images? How far appart are these images in time?

Looks like I will be clouded out tonight (taking one or two images is typically no probelm for me - it's just a matter of entering the co-ordinates into my automated observing plan) so I can't look at your co-ordinates.

Now you do realise that images (film, scanned film and CCD) all suffer from flaws (film flaws, dust, dirt and scratches, cosmic ray hits, read noise and hot pixels). Investigating every 'disappearing' dot based on a single image is really a waste of time. A more reliable method is observing a moving dot and even that can be misleading (but far moe reliable that a single disappearing dot)

Cheers

[Joe Keller]

I found Barbarossa on a third archive plate. There is no longer any doubt of Barbarossa's reality nor of Barbarossa's position. Using the two most credible of these "disappearing dots" (1987 & 1997) Barbarossa's period, assuming circular orbit, matches the progression of the imperfect 5:2 Jupiter:Saturn resonance, to 3% accuracy. As in earlier estimates, Barbarossa aligns in longitude, with one of the five (mean, corrected for Jupiter & Saturn eccentricity) resonance points, to 0.4 or maybe 0.1 deg error.

Corrected for Earth parallax, the four points (there are two competing dots on the 1954 plate scan) lie nearly on a great circle. The change in angular speed corresponds to orbital eccentricity of at least 0.1, or at least 0.25, if one or the other of the 1954 points is used in addition to the 1987 & 1997 points. The Red comparison-based magnitudes of three of the objects are all +17.6 +/- 0.3; one of the 1954 objects is about Red +18.3.

The plates (from online 1.0"-resolution scanned versions) are:

1. POSS-I E (a.k.a. POSS-I Red)(exposures for this series varied from 2400 to 4200 sec) Plate XE671, February 25, 1954, epoch 1954.154.

First dot: RA 11h 03m 12.4s Decl -5deg 58' 09"

I determined the Red magnitude as +17.6 by comparison with the USNO-B Red1 magnitude (Red1 was chiefly determined from plates of this series) of a nearby star with stable magnitude. I found this dot, March 29. If it is Barbarossa, then Barbarossa's eccentricity must be at least 0.25, assuming the validity of the 1987 & 1997 dots.

Second dot: RA 11h 02m 25.16s Decl -5deg 56' 11.3"

By comparison with nearby stars, this dot's Red mag is about +18.3. I found this dot, March 28. It is the brightest of a "flying circus" of five disappearing dots spread over about 1'. Together with the 1987 & 1997 dots, it would imply an eccentricity of at least 0.1.


2. SERC-ER (a.k.a. SERC2 Red)(exposure 3600 sec), Plate 713, January 31, 1987, epoch 1987.08215.

RA 11h 18m 03.18s Decl -7deg 58' 46.1"

Because this sky survey was only one of three used to determine the USNO-B Red2 magnitudes, I determined the magnitude of Barbarossa on this plate, by comparison with both the R1 & R2 magnitudes of four nearby stars, finding +17.3. I saw Barbarossa on this plate, March 4, and realized on March 5 that what I saw, was Barbarossa.


3. SERC-I (a.k.a. Optical Infrared)(exposure 5400 sec), Plate IS713(A438), March 3, 1997, epoch 1997.16711.

RA 11h 22m 16.77s Decl -8deg 29' 30.9"

I determined Barbarossa's Infrared magnitude as +18.1 by comparison with two nearby stars. Though the authors of the USNO-B catalog warn that it is a relatively inaccurate source for magnitudes of bright stars, I used the USNO-B's I-R value for Capella, 0.2, to correct the sunlit Barbarossa's Red magnitude to +17.9. I found this dot March 31. I've found no other Optical Infrared plate online with which to prove the disappearance of this dot. Instead, I found that it is absent from both the SERC Red and MASS IR J,K,H plate scan series, indicating, if not disappearance, then an aberrantly narrow spectrum.


The 1987-1997 track implies a 2775 year period for circular orbit. Recent values of Jupiter's and Saturn's periods indicate that their 5:2 resonance progesses with a period of 2696 yr.

Corrected for April 1 Earth parallax, Barbarossa's geocentric coordinates tonight, assuming a circular orbit through the 1987 & 1997 objects, are:

RA 11h 26m 30.9s Decl -9deg 00' 11"

The position might be 7' W to 1.5' E of this, if one or the other of the 1954 dots is used for prediction instead of, or in addition to, the 1997 dot. Last night Steve Riley imaged an approx. mag. +17.3 dot which tonight will be 3' NW of these coordinates (only 1' above the predicted track), if Steve indeed imaged Barbarossa.

Barbarossa's estimated apparent diameter is 0.9". Barbarossa's estimated retrograde motion is 0.7"/hr.

Sincerely,
Joseph C. Keller, M. D.

[higginsdj]

When you say 'disappearing dot' do you mean a single pixel? If you are then I am afraid you are mistaken in your identification of this object. On those plates a mag 17-18 object will occupy multiple pixels ('circles' comprising no less than 8-12 pixels)

Single 'brightish' pixels are noise that have not been fully removed from the image.

Edit: I ran your positions through some orbit calculators and I should point out that the 3 positions you indicate cannot fit an orbit of any type.

Cheers

[Joe Keller]

Six of the seven "disappearing dots" I've found (on online archive sky survey plate scans) with magnitudes about +18.8 or brighter, fit a mildly elliptical two-body mutual orbit with major axis at least 1.95 AU. The eccentricity is between 0.19 (absolute lower bound) and about 0.38 (better than 67% confidence upper bound). The mutual orbital plane is inclined about 10.7deg to Barbarossa's orbital plane. The mass ratio of these two, Barbarossa and Frey, is 5:1. The period is 42 yr. The implied mass of Barbarossa is between 0.0051 solar mass for 0.19 eccentricity and 0.0080 solar mass for 0.38 eccentricity. The orbital period of the center-of-mass about the sun is 2850 yr, assuming a circular orbit.

The magnitudes of Barbarossa are about +17.3, 17.9 & 18.0; of Frey, +17.6, 18.3 & 18.8. The dimmest Frey magnitude occurs, near maximum elongation, but much nearer to a point at which Frey's orbit intersects that plane (through the center of mass of Barbarossa & Frey) which is parallel to the principal plane of the solar system.

There might be a dust belt there. Neither Frey nor Barbarossa could be found on the March 1986 plate. Then, the line through Barbarossa & Frey, as seen from Earth, was theoretically only 1.8deg from the plane which contained the center of mass and was parallel to the principal plane of the solar system. Now, 21 yrs = 0.50 orbit later, the same situation holds.

In this model, the geocentric coordinates for April 5 (for the next few days the usual minus 1s RA & +6" Decl per day correction applies) are:

Barbarossa: RA 11 26 07.5 Decl -8 59 53.5

Frey: RA 11 27 35.5 Decl -9 10 47.5

[CometGuy]

Joe,

I am with David on this one. Regarding Joan Genebriera's image posted on metaresearch.org all I see is a processing defect.

Terry

[higginsdj]

This is like one of those SPAM posts with randomly selected words strung together to sound something like something but ends up being gibberish!

Cheers

[Joe Keller]

I continue to work on arranging the "disappearing dots" into a model of Barbarossa's system. Herein I present my latest, most accurate (correcting my transcription error in the coordinates of one object, and increasing the accuracy of my adjustments for Earth parallax), most plausible and most conservative model to date.

Three dots are plausibly Barbarossa. These are (using the names of convenience I've assigned them as I've worked):

A2. POSS-I (Red) plate date 1954.154, geocentric position RA 11 02 25.16 Decl -5 56 11.3

C. SERC (Red) plate 1987.08215, RA 11 18 03.18 Decl -7 58 46.1

D. SERC-I (Optical Infrared) plate 1997.16711, RA 11 22 16.77 Decl -8 29 30.9.

The position of C differs 95 arcsec from its expected, parallax-corrected, great-circle interpolated position between A2 & D. Though this deviation is, I think, about 10x bigger than the errors inherent in my model or in my calculations, the only component of the deviation big enough to demand explanation, is that perpendicular to Barbarossa's path (i.e., the path from A2 to D).

Two more dots are consistent with the moon Frey in a near-circular approx. 1.4 AU, 22-yr orbit inclined only a few degrees to Barbarossa's orbital plane:

A. POSS-I (Red) plate date 1954.154, geocentric position RA 11 03 12.4 Decl -5 58 09

D2. SERC-I (Optical Infrared) plate 1997.16711, RA 11 22 32.9 Decl -8 26 56.

I've found yet three more dots, two of which could be consistent with the moon Freya in a near-circular 2 AU orbit moderately inclined to Frey's:

B8. UK-Red plate approx. date 1986.199, geocentric position RA 11 14 58 Decl -7 42 20

or

C5. SERC (Red) plate 1987.08215, RA 11 16 04.4 Decl -7 47 51 ;

and

E2. SERC-I (Optical Infrared) plate approx. date 1995.140, approx. RA 11 19 43 Decl -8 06 50.

These orbits imply about 0.0054 solar mass for Barbarossa and much lower mass for Frey and Freya. The total system mass thus could be close to the earlier predicted 0.0068 solar mass which smoothed the net Pioneer 10/11 accelerations.

[Joe Keller]

The words aren't random and it's not gibberish. It might not be correct, but it's not gibberish. You know that. Why aren't you telling the truth?

[higginsdj]

<<Text removed to avoid the possibility that it might be construed as ridicule>>

Cheers

[bloodhound31]

Should I stick my nose in here?

Boys, be nice.

Lets not let this turn nasty. Joe, continue your search for Barbarossa. You may be wasting your time, you may not. I hope you find it. Anyone who doesnt want to look for this possible object, no one is holding a gun to your head. Let Joe and his believers search. He's not hurting anyone.

My two bob's worth.

Baz.:D

[higginsdj]

Don't get me wrong. I do believe there are lots of objects out there yet to be discovered and I will be the first to apologise to Joe if he actually does find something (that can be proven by scientific methods and/or direct observation) - I just believe in proper observation methodologies and the scientific method - neither of which has been publicly displayed by Joe to date.

And lets not keep calling it Barbarossa. The name belongs to Minor Planet 1860, discovered by P Wild at Zimmerwald on 28 Sept 1973 and the IAU will not let you re-use it or anything resembling it.

Cheers

[duncan]

G,day all,

It amuses me no end to watch all this bickering over something that might or might not be out there. Surely if there was anything to this then the really big scopes on Earth and Hubble would have had a go.:shrug: !!!
All this argueing won't solve anything. Put something big really BIG on it to try and solve it once and for all.

Just my 2cents worth.
Cheers,
Duncan:thumbsup:

[higginsdj]

Not being a mathematician myself, a thought just occurred to me......and I could very well be wrong..... but can't I fit 1, 2 or 3 points to any curve ie I could pick 3 random points and I could fit it to any curve/orbit I choose?

With such short arcs over such a long time span how much impact would timestamps have on this situation?

Cheers

[Joe Keller]

Here's the latest "dot theory". These three dots are Barbarossa:

A2. POSS-I (Red) plate date 1954.154, geocentric position RA 11 02 25.16 Decl -5 56 11.3

C3. SERC (Red) plate 1987.08215, RA 11 18 37.6 Decl -7 54 09.5

D. SERC-I (Optical Infrared) plate 1997.16711, RA 11 22 16.77 Decl -8 29 30.9.

These three dots are Frey:

A. POSS-I (Red) plate date 1954.154, geocentric position RA 11 03 12.4 Decl -5 58 09

C. SERC (Red) plate 1987.08215, RA 11 18 03.18 Decl -7 58 46.1

D2. SERC-I (Optical Infrared) plate 1997.16711, RA 11 22 32.9 Decl -8 26 56.

(Some of the other dots are Freya. There are some disappearing dots on the 1986 and 1995 plates which could be these bodies too.)

I assumed that A2 & D are Barbarossa, then making my most accurate correction for Earth parallax, interpolated the expected position for Barbarossa on the 1987 plate. Both C and C3 are a small distance away from that position.

Then I drew lines between C3 & C, A2 & A, D & D2. If these are Barbarossa & Frey in mutual orbit, the center of mass should be displaced at a constant rate. This is best seen by graphing all six bodies on the same sheet, each body relative to the presumed Barbarossa of the pair for its epoch. Generally there will be one mass ratio which makes the centers of mass collinear.
However, when the centers of mass became collinear, they also assumed the correct distance ratio, i.e., constant speed, to within 2% accuracy. (I refer to the residual small speed remaining after the speed from A2 to D is deducted.) This is a very precise and unlikely result. The implied period for circular orbit around the sun was 2847 yr (vs 2688 yr for the J:S resonance progression). Furthermore the mass ratio which gave this precisely constant-velocity center of mass, was 1:1. The conditioning of the graphical solution was such that a 1.2 :1 ratio either way might occur, but certainly not 1.5 :1. Alpha Centauri A & B are said to have a 1.2 :1 ratio, as do Earth & Venus.

The mutual orbit cannot be perfectly circular, because no ellipse centered on the center of mass, fit the points. Slight displacement of the ellipse center (if a noncircular orbital ellipse is tilted, the center of mass generally is not even a focus) allows an infinitude of ellipses. I chose one such that was especially easy to calculate, and found constant angular speed between A, C & D, within 10%; distance between Barbarossa & Frey, 0.7 AU; inclination 18 deg; tilt to Barbarossa's solar orbit, 25 deg; tilt to orbital plane, 30.5 deg; combined mass of Barbarossa & Frey, 0.0036 solar masses.

The trajectory of the presumed center of mass of Barbarossa & Frey, is so constant that Freya likely would have to be of much smaller mass than Frey, or much more distant from Barbarossa. Alternatively, let 1954 be t=0, 1997 be t=1. The midpoint of the interval A-C then is t=3/8 and the midpoint of C-D is t=7/8. If Freya were at conjunction (near our line of sight to the center of mass of Barbarossa & Frey) at t=5/8, then to a first-order approximation the acceleration due to Freya would be zero in the plane of the celestial sphere.

In this model, Barbarossa & Frey are always within 15 arcminutes of the more recent of my various predicted positions. So, the best I have to offer now, is to keep looking within 15' of those coordinates.

[Joe Keller]

I've spent much of my time this week accurately confirming that J. Genebriera did indeed image Frey on March 25, 2007 (00:42 UT) and that S. Riley did indeed image Barbarossa on April 1, 2007 (07:39 UT). The J2000 coordinates of Genebriera's object are

11 26 22.2 -9 4 59

and of Riley's

11 26 25.0 -8 57 26.

(I can't access Aladin from this library so the Riley coordinates are from measuring on the screen with a ruler and therefore slightly rough; 3mm on the screen is 1s RA.)

The center of gravity slowed only 0.69% between the second and third segments, i.e., [Objects C3 & C 1987, Objects D & D2 1997] and [Objects D & D2 1997, Riley & Genbriera Objects 2007]. This corresponds to about 2s RA. The direction changed only 0.0046 radian (i.e., 0.46%) between the two segments. These deviations are smaller than likely would arise from observation bias: I searched entire 15x15' square images for "disappearing dots" and parts of adjoining squares also, rarely finding more than one or two starlike "disappearing dots" per square.

The direction changed 0.86% between the first and second segments, i.e., [Objects A2 & A 1954, Objects C3 & C 1987] and [Objects C3 & C 1987, Objects D & D2 1997]. The center of gravity slowed 3.3%.

The correction for Earth parallax was made by interpolating the sun's position according to old volumes of the American Ephemeris & Astronomical Almanac in the Iowa State Univ. library. For 1954 I had to use the formula therein to convert to J2000 coordinates. On an IBM486 computer I wrote a "BASIC" program to find the rectangular coordinates of every object precisely. I adjusted the objects' distance from the sun (presumed the same for all) so that the angle subtended between 1954 and 2007 equalled that for a body with a slightly elliptical orbit of period 2688 yr (my best estimate of the period of progression of the 5:2 Jupiter:Saturn resonance) when at said distance from the sun.

Then I adjusted the mass ratio of Barbarossa (i.e., A2, C3, D & Riley) and Frey (i.e., A, C, D2 & Genebriera) to 0.62:0.38, at which the torsion of the great circle was about constant: that is, the (small) break between the first (32.9 yr) and second (10.1 yr) segments was about twice the break between the second (10.1 yr) and last (10.1 yr) segments.

Corrections for proper motion of reference stars, between the 1987 date of the SERC-Red reference plate, and 2007 or 1954, all were negligible (1997 had its own Aladin reference plate). The aberration of light from the objects is negligible because it affects the reference stars as well; the aberration of sunlight is negligible because the sun's position is needed only for the small Earth parallax correction. The correction for the different distances of Barbarossa vs. Frey from the sun, is negligible.

The decreasing angular speed could be due to the influence of a small distant moon. The moon(s) seem to orbit Barbarossa in a plane near that of Barbarossa's orbit, so any torsion of the great circle would be relatively small.
The strange shapes of both Genebriera's and Riley's objects might be due to rings like Saturn's. The Roche limits for these bodies, a likely distance for rings, would be 1-2" depending on densities. The bodies' diameters should be about 0.8". A barely detectable planetary disk could make a spot smaller than that of a star of equal bolometric magnitude.

[GeoffW1]

David,

Very rough stuff this, could be better put.

I am way way behind both of you except on one thing: before assuming the position of resident sceptic (and there is nothing wrong with doubting) please spell it correctly "scepticism".

GeoffW1

[higginsdj]

OK, based on some physics and what I can gather from your information:

* Semi Major Axis for the parent object = 193au
* Assuming Moon and Planet have same density the Roche Limit of 281,000 km (your 2" estimate of the ring system) means a planet with a radius of 223,016km (3 times the radius of Jupiter) - I'm aware it's a bit of a stretch to assume that an object of this size would have the same density as it's moon(s) so this is just a ballpark figure.
* Assuming that the object does not generate any light itself and based on a low albedo (same as an Asteroid) then an object this size will have an absolute magnitude of -9.1 and at 193AU, an apparent brightness of mag 2.7!

Can you account for why an object so large and relatively close has had no impact on the orbits of 800 currently known TNO's?

Given that the seeing at the 2 sites from which you obtained the reference images are likely to be 3"-4" (regardless of what their pixel scale is) what size does the actual image measure for this planetary disk?

Some advice on measuring an image and detecting an object. A 14" at f/10 is likely to produce a pixel scale of no finer than 0.5"/pixel. However, given the seeing limit (mentioned previously), an object is not going to occupy less then Seeing limit divided by 0.5 pixels in X and Y directions (for my 14" scope example). This is a physical limit. Anything smaller than this, particularly with a sharp increase in brightness to adjacent pixels is not a real object no matter how conveniently located on an image.

Cheers

[Joe Keller]

"Dot theory" update: high significance.


With a more accurate calculation of Earth parallax, the above-mentioned alignment of presumed centers of gravity of A (1954), C (1987) & D (1997) objects became less perfect. I've been resorting to the IBM486 computer more, to avoid such inaccuracies.

Also, I've found two "disappearing dots" of interest on the B (1986) plate:

"B3" 11 16 51.55 -7 49 41.1
"B" 11 16 56.07 -7 55 14.3.

The following theory is hindered by my lack of success in finding accurately corresponding objects on the C plate. However A2 & A, B3 & B, and J. Genebriera's (March 25) & S. Riley's (April 1) objects (see above for coordinates and other details) seem to correspond to Barbarossa & Frey, resp.

I found accurate heliocentric coordinates for these six objects, using old ephemerides and the above-mentioned "BASIC" computer program. I assumed that all six objects have the same distance from the sun. This distance, 197.7283 AU, was adjusted, as above, so that the overall angular speed 1954-2007 (equivalent to a circular orbit with period 2811.866 yr) would equal that for an elliptical orbit with period 2688.000 yr.

The mass ratio of the contemporaneous pairs of objects was adjusted so that the two great-circle arcs, from A to B and from B to Genebriera/Riley, had exactly the same direction to within a millionth of a radian. The resulting mass ratio was 0.87710:0.12280. Generally there is a ratio which will cause the directions to be the same, because often an equation will have a real solution. Generally a different equation will not be simultaneously satisfied. Here, I found simultaneously that the angular speeds from 1954 to 1986 and from 1986 to 2007 became practically equal, only 0.0307% different (0.00116% per yr).

I checked this graphically as above. By successive approximations at the computer, I found the geocentric coordinates (for G.'s & R.'s objects, coordinates always were adjusted to the midpoint of Genebriera's & Riley's observation times) which would put Genebriera's object on a perfect constant-speed heliocentric great circle with objects A2 & B3. Then I graphed all six objects on the same sheet, with these three points superimposed as the origins for objects of their epoch. I approximated the short great circles between contemporaneous points, as lines, and neglected the nonparallelism of longitude lines. (Also on my graph I neglected the 1% difference between RA & Decl degrees, at this Declination, about -7.2 deg ave.)

When I found the centers of gravity according to the above ratio, they were, as expected, collinear and showing constant speed. (On the graph paper, the errors of the collinearity and the constancy of speed both were about 1%.) Translating the Genebriera & Riley objects equally so that Genebriera's object also lay on the origin, I found by quick but generalized searching at the computer, the approximate unique ellipse centered at that origin and passing through the three presumed Frey objects (A, B, and Riley's). The ellipse was tipped 10deg NE-SW (i.e., 37deg to Barbarossa's orbit). The major and minor semiaxes were 0.25deg (0.86 AU) and 0.10deg, resp., giving i=22deg and a total tip for the Barbarossa-Frey system, of 43deg to Barbarossa's solar orbit.

The positions on the presumed orbital circle then were found. This was consistent with an orbital period of 8.3 yr (almost 4 revolutions, 1954-1986, & almost 2.5, 1986-2007) with only 3% discrepancy between the A-B and B-Riley arcs. The total mass Barbarossa+Frey would be 0.0094 solar mass (Barbarossa alone, 0.00825 solar mass).

The theoretical mass at this distance, needed to produced the CMB dipole, was calculated as above using precise 200- or 2000-step trapezoidal rule integration, and found to be 0.0116 solar mass. (Any additional bodies in the Barbarossa-Frey system, unless closely orbiting Barbarossa or Frey, or very distant, would need to have small mass because of the precise motion of the Barbarossa-Frey center of mass).

An inaccuracy in the tidal Pioneer 10/11 acceleration calculation above, was fixed, and the new parameters applied. The net sunward anomalous Pioneer 10/11 acceleration, after subtracting the tidal forces from the presumed 0.0116 solar mass Barbarossa system, at the four distances tabulated by O. Olsen (A&A, op. cit. 2007 above) becomes:

27 AU: 6.03*10^(-8) cm/s^2
45: 5.38* "
52: 5.68* "
63: 4.25* "

The best figure for the Hubble shift is 72 km/s/Mpc, which is equivalent to 7.0*10^(-8) cm/s^2. Assuming that near the sun, the net anomalous acceleration equals the Hubble-equivalent acceleration (the Galileo and Ulysses measurements of the anomalous acceleration lack the accuracy to confirm or refute this) the anomalous acceleration approximates a normal distribution with peak 7.0 and standard deviation 53 AU:

observed/predicted/difference (relative scale, i.e., 7.0*10^(-8) is 1.00)
0 AU: ~1.0+-0.3/1.00/?
27: 0.86/0.88/-0.02
45: 0.77/0.70/0.07
52: 0.81/0.61/0.20
63: 0.61/0.50/0.11

So, accounting for the tidal force of Barbarossa et al, the magnitude of the anomalous acceleration, as a function of radius, approximates a normal curve with standard deviation roughly 53 AU; with a small hump added near 53 AU.

Object A2 has four dimmer disappearing dots within about an arcminute (6,000,000 mi at 198 AU); two of these are a pair about 20" (2,000,000 mi) apart. Likewise object B3 has a pair of dimmer disappearing dots about 0.75' away and about 10" apart. Barbarossa is a cold brown dwarf: these might be Barbarossa's inner planets. Frey is Barbarossa's Jupiter. Barbarossa's next-biggest satellite already has been named Freya.

Five more starlike disappearing dots have been found (on the C, and D 1997 Optical IR) sky survey images, which seem bright enough to be Barbarossa, Frey or Freya and are close enough to fall on the same sheet of graph paper on which I graphed the Barbarossa-Frey orbit. The largest telescopes used so far to search prospectively for Barbarossa or Frey have been the 16" telescopes of J. Genebriera and S. Riley. Comparison with sky surveys shows that the sky survey dots I seek to relocate are near, if not beyond, the detection limits of these telescopes. Genebriera and Riley have achieved, so far, one detection apiece (Barbarossa and Frey, resp., according to this latest theory).

[astroron]

:hi: Joe, could you please explain why the rest of the Planetary Astronomers are not acknowledging your discovery:shrug:

[CometGuy]

That seemed a simple enough question that David posed...at 0.01 Solar masses it should be quite bright even at 190 AU. Even Jupiter at 0.001 solar masses, would be magnitude 12-13 at that distance. How it could be at the limit of a 16" telescope + CCD camera beats me. This would indicate something closer to earth size.

I also I see one of the most experienced amateur astronomers on this list (David Higgins) being told off for challenging this claim and giving good reasons.....come on guys !

Terry

[higginsdj]

It's OK Terry. I believe Geoff is a teacher and teachers always pick up the spelling mistakes - and rightly so. :)

My only concern here is that others may 'blindly' take up the search based on Joes data, spending money with online remote scopes (as they appear to have done in the Meta Research forum) looking for something that does not appear to be based on practical physics, current theory or good observational principals.

I've put up some alternative views based on current theory and physics - people can make up their own minds. (Note that Meta Research is based on principals of Alternative Theory). Can you imagine, he's claiming a whole second solar system just 193 AU from our own, protoplanetary disc and all!

Cheers

[higginsdj]

The practical detection limit for a 0.4m scope and CCD is around R magnitude 22. You will find that many 0.35m scopes + CCD's have been regularly able to detect moving targets at R Mag 21 in the UK, US and Australia. Of course this will vary with sky brightness (and I am not talking about darks sky limits here)

Cheers

[Joe Keller]

(Moderator: for the historical record, please do not delete my previous post. It contains only text and therefore consumes little memory.)

"Dot theory" update: high significance.

With a more accurate calculation of Earth parallax, the above-mentioned alignment of presumed centers of gravity of A (1954), C (1987) & D (1997) objects became less perfect. I've been resorting to the IBM486 computer more, to avoid such inaccuracies.

Also, I've found two "disappearing dots" of interest on the B (1986) plate:

"B3" 11 16 51.55 -7 49 41.1
"B" 11 16 56.07 -7 55 14.3.

The following theory is hindered by my lack of success in finding accurately corresponding objects on the C plate. However A2 & A, B3 & B, and J. Genebriera's (March 25) & S. Riley's (April 1) objects (see above for coordinates and other details) seem to correspond to Barbarossa & Frey, resp.

I found accurate heliocentric coordinates for these six objects, using old ephemerides and the above-mentioned "BASIC" computer program. I assumed that all six objects have the same distance from the sun. This distance, 197.7283 AU, was adjusted, as above, so that the overall angular speed 1954-2007 (equivalent to a circular orbit with period 2811.866 yr) would equal that for an elliptical orbit with period 2688.000 yr.

The mass ratio of the contemporaneous pairs of objects was adjusted so that the two great-circle arcs, from center-of-mass A to center-of-mass B and from c.o.m. B to c.o.m. Genebriera/Riley, had exactly the same direction to within a millionth of a radian. The resulting mass ratio was 0.87710:0.12280. Generally there is a ratio which will cause the directions to be the same, because often an equation will have a real solution. Generally a different equation will not be simultaneously satisfied. Here, I found simultaneously that the c.o.m. angular speeds from 1954 to 1986 and from 1986 to 2007 became nearly equal, only a -0.0508% change (-0.001915% per yr). This is typical of the speed change expected in the lower range of possible solar orbital eccentricities.

I checked this graphically as above. By successive approximations at the computer, I found the geocentric coordinates (for G.'s & R.'s objects, coordinates always were adjusted to the midpoint of Genebriera's & Riley's observation times) which would put Genebriera's object on a perfect constant-speed heliocentric great circle with objects A2 & B3. Then I graphed all six objects on the same sheet, with these three points superimposed as the origins for objects of their epoch. I approximated the short great circles between contemporaneous points, as lines, and neglected the nonparallelism of longitude lines. (Also on my graph I neglected the difference between RA degrees & Decl degrees: this changes 0.7% between 6 & 9deg Decl.)

When I found the centers of gravity according to the above ratio, they were, as expected, collinear and showing constant speed. On the graph paper, the errors of the collinearity and of the constancy of speed both were about 1% of the graphed portion, which was in turn about 1% of the total motion; i.e., about 10^(-4) accuracy overall. Translating the Genebriera & Riley objects equally so that Genebriera's object also lay at the origin, I found by quick but generalized searching at the computer, the approximate unique ellipse centered at that origin and passing through the three presumed Frey objects (A, B, and Riley's). The ellipse was tipped 10deg NE-SW (i.e., 37deg to Barbarossa's orbit). The major and minor semiaxes were 0.25deg (0.86 AU) and 0.10deg, resp., giving i=22deg and a total tip for the Barbarossa-Frey system, of 43deg to Barbarossa's solar orbit.

The positions on the presumed orbital circle then were found. This was consistent with an orbital period of 8.3 yr (almost 4 revolutions, 1954-1986, & almost 2.5, 1986-2007) with only 3% discrepancy between the A-B and B-Riley arcs. The total mass Barbarossa+Frey would be 0.0094 solar mass (Barbarossa alone, 0.00825 solar mass).

The theoretical mass at this distance, needed to produced the CMB dipole, was calculated as above using precise 200- or 2000-step trapezoidal rule integration, and found to be 0.0116 solar mass. (Any additional bodies in the Barbarossa-Frey system, unless closely orbiting Barbarossa or Frey, or very distant, would need to have small mass because of the precise motion of the Barbarossa-Frey center of mass).

An inaccuracy in the tidal Pioneer 10/11 acceleration calculation above, was fixed, and the new parameters applied. The net sunward anomalous Pioneer 10/11 acceleration, after subtracting the tidal forces from the presumed 0.0116 solar mass Barbarossa system, at the four distances tabulated by O. Olsen (A&A, op. cit. 2007 above) becomes:

27 AU: 6.03*10^(-8) cm/s^2
45: 5.38* "
52: 5.68* "
63: 4.25* "

The best figure for the Hubble shift is 72 km/s/Mpc, which is equivalent to 7.0*10^(-8) cm/s^2. Assuming that near the sun, the net anomalous acceleration equals the Hubble-equivalent acceleration (the Galileo and Ulysses measurements of the anomalous acceleration lack the accuracy to confirm or refute this) the anomalous acceleration approximates a normal distribution with peak 7.0 and standard deviation 53 AU:


observed/predicted/difference (relative scale, i.e., 7.0*10^(-8) is 1.00)

0 AU: ~1.0+-0.3/1.00/?
27: 0.86/0.88/-0.02
45: 0.77/0.70/0.07
52: 0.81/0.61/0.20
63: 0.61/0.50/0.11

So, accounting for the tidal force of Barbarossa et al, the magnitude of the anomalous acceleration, as a function of radius, approximates a normal curve with standard deviation roughly 53 AU; with a small hump added near 53 AU.
Object A2 has four dimmer disappearing dots within about an arcminute (6,000,000 mi at 198 AU); two of these are a pair about 20" (2,000,000 mi) apart. Object B3 has probably one dimmer disappearing dot about 45" S. Barbarossa is a cold brown dwarf: these might be Barbarossa's inner planets. Frey is Barbarossa's Jupiter. Barbarossa's next-biggest satellite already has been named Freya.

Five additional more-or-less starlike disappearing dots have been found (on my "C", 1987 Red, and "D" 1997 Optical IR) sky survey images, which seem bright enough to be Barbarossa, Frey or Freya and are close enough to fall on the same sheet of graph paper on which I graphed the Barbarossa-Frey orbit. The largest telescopes used so far to search prospectively for Barbarossa or Frey have been the 16" telescopes of J. Genebriera and S. Riley. Comparison with sky surveys shows that the sky survey dots I seek to relocate are near, if not beyond, the detection limits of these telescopes. Genebriera and Riley have achieved, so far, one detection apiece (Barbarossa and Frey, resp., according to this latest theory).

[Joe Keller]

"Dot update"

One of my five additional nearby starlike disappearing Sky Survey dots, is

"C11" (Jan 31 1987, SERC Red) Strasbourg "Aladin"
RA 11 18 00.41 Decl -8 01 57.7.

Assuming this is Frey and that my previous post is valid, I find with my IBM 486 "BASIC" program above, that Barbarossa should be on this 1987 plate at

11 18 03.006 -7 56 29.4.

There is a magnitude +19 star at

11 18 02.80 -7 56 21.8.

This star seems brighter relative to its neighbors, on this 1987 La Silla sky survey vs. the 1986 "UK Red" sky survey. Its R1 & R2 USNO-B catalog magnitudes are nearly equal, but the documentation doesn't reveal the relative influence of the two or three plates used to compute each magnitude. So, it might be a nonresolved conjunction, or (less likely) occultation.

My computerized Earth parallax correction uses the sun's apparent position. Barbarossa orbits the center of gravity of the solar system: including Jupiter & Saturn. Saturn was near quadrature with Barbarossa then, so its influence changed little between 1986 & 1987. On the other hand, Jupiter was nearly opposite Barbarossa. Jupiter's 30 degree motion in one year would subtract about 0.15s RA from Barbarossa's predicted apparent position in 1987, vs. 1986. So, only the predicted and observed Declinations of the star are significantly discrepant.

My very first discovered sky survey "Barbarossa" object, "C", lies 145" exactly S of the predicted position of Barbarossa. My pixel analysis of this object indicated that it is at least two different unresolved objects (the center of luminosity doesn't lie in the brightest pixel). Object C might be an unresolved conjunction of two of Barbarossa's inner planets. (Such "inner planets" appear on the A and B plates.) If so, then these planet(s) would need 6% the mass of Barbarossa, to move the center of gravity of the non-Frey, inner, portion of Barbarossa's system, 145*0.06/1.06 = 8" S. This would place the predicted Barbarossa at the coordinates of the observed star.

[Joe Keller]

Steve Riley (Buena Vista Observatory, California) photographed what I call his "Object #1" at approx. 07:39 UT on April 1, 2007. By comparison with the 1987 La Silla archive image (on which it is absent), its position (J2000 geocentric coordinates) is

RA 11 26 24.6 Decl -8 57 48.5.

Riley has a new photo which shows what I call his "Object #2" likewise not found on the 1987 archive image. It was taken (midrange of contributing exposures) 05:41 UT on April 24, 2007. Its coordinates (again by comparison with 1987) are

RA 11 26 09.58 Decl -8 55 56.0.

(This second point is corrected for the proper motion of the reference stars but the first point isn't. That's alright, because the relevant stellar PMs are uncertain and of marginal size in both cases, especially the first.)

Using the ephemeris position of the sun at these times, in my above-mentioned computer program, I found that the heliocentric xyz position of these objects differs as though they are the same object which has moved 0.1 AU prograde in an orbit inclined 34 degrees (and moving south) to the celestial equator. Barbarossa's orbit would move it 0.03 AU prograde and is inclined 27 degrees.

There is little remaining discrepancy in the x coordinate. Two-thirds of the remaining discrepancy in the y coordinate is removed by employing my above-mentioned model of the Barbarossa-Frey orbit, but this model does not remove the z coordinate discrepancy. If I do not insist on a circular actual orbit for Frey about Barbarossa, I can move the apparent semimajor axis 37 degrees, until it is parallel to Barbarossa's solar orbit: then, the predicted x, y, and z coordinates of Riley's Object #2 all become roughly what is observed.

Riley uses an 11" telescope [not 16" as I had assumed earlier!]. In Riley's photo I found two different +19.3 stars with stable catalog magnitudes. One of these stars was obvious, and one amounted to a barely discernible pixel overdensity. So, I think the position of Riley's Object #2 is more important than its appearance, which is faint and abnormally small.

Most of the apparent motion between Riley's Objects #1 & #2, is due to Earth parallax. The daily change in Earth parallax is shrinking rapidly, but for the next few days, linear interpolation will predict the geocentric coordinates accurately enough to find Riley's Object, which I think is Frey. When Riley's Object (Frey) is found a third time, then quadratic interpolation can be used to predict the geocentric coordinates.

[Joe Keller]

open letter to the Secretary of the Navy (USA)

To: the Hon. Donald C. Winter, Ph.D.
Secretary of the Navy
Washington, D. C., USA

From: Joseph C. Keller, M. D.
POB 9122
Ames, Iowa 50014 USA

Date: May 15, 2007

Subject: planetary discovery

Dear Secretary Winter:

Please order someone with a big telescope to look for planets near these coordinates:

RA 11h 26m Decl -8.9 deg (J2000)

At 00:42 on March 25, 2007, UT, amateur astronomer Joan Genebriera of Barcelona, Spain, using a 16” telescope and electronic camera at Tacande Observatory on Tenerife in the Canary Is., aiming her telescope at coordinates I provided, recorded objects near

RA 11 26 22.2 Decl -9 04 59 and
RA 11 26 31.8 Decl -9 00 11

These objects seem to be identical with objects found on the online scan of the1954 (48” Schmidt camera) Palomar sky survey plate “POSS-I Red” at these coordinates:

RA 11 02 25.2 Decl -5 56 11 and
RA 11 03 12.4 Decl -5 58 09

and the scan of the 1986 (also 48” Schmidt) Australian sky survey plate “UK Red” at:

RA 11 16 51.8 Decl -7 49 40 and
RA 11 16 56.1 Decl -7 55 14

Assuming a mass ratio of 0.865: 0.135, and a distance, from the sun, of 197.75 Astronomical Units, the center of gravity seems to be in orbit around the sun. The eccentricity of the orbit is < 0.009 with 90% confidence. (The eccentricity of Neptune’s orbit is 0.009.) Though uncommon, double star orbits of this size and eccentricity are known. Known double stars are observationally biased toward systems of more equal mass.

The apparent period of the orbit equals the period of advancement of the 5:2 Jupiter:Saturn resonance, to the accuracy to which the latter is known. Correction for the eccentricities of Jupiter’s and Saturn’s orbits shows that my newly discovered objects lie, in projection, within a fraction of a degree of one of the five mean Jupiter:Saturn resonance points.

The objects lie only two degrees prograde from the positive Cosmic Microwave Background dipole. The dipole lies on the objects’ orbit to within a fraction of a degree, the accuracy to which the dipole is known. This is consistent with a new, gravitational theory of the CMB, which would revolutionize our understanding of the relationship between gravity and electricity.

For reasonable masses, correction for the tidal gravity of the objects, reduces the variation of the Pioneer Anomalous Acceleration. The net Anomalous Acceleration becomes fairly smoothly decreasing with distance from the sun.

I think that despite the likely 0.01 solar mass for the combined objects, they fail to disrupt the solar system, because small shifts in the orbital planes of the known planets, counter the torque. Depending on composition, the objects might have smaller diameters than predicted by published giant planet theory.

The first photos taken in the search for these objects were by amateur astronomer Robert Turner of England, using the 14” telescope and electronic camera of the Bradford College Observatory, also on Tenerife, Canary Is. Amateur astronomer Steve Riley of California, USA, took many photos using an 11” telescope and electronic camera at Buena Vista Observatory in California. The objects are near the detection limit for all these observers. Known stars of magnitude equal to the objects, often are absent or distorted in the photos. Despite these problems, Turner’s later photos, Genebriera’s, and especially Riley’s, show several alternative or additional detections of the objects.

Sincerely,

Joseph C. Keller, M. D.

[higginsdj]

Now Joe you've only asked him to point to one set of co-ordinates yet you believe your discovery could be any any one of 5! Wouldn't it have been prudent to ask for all 5 locations to be searched?

Also, how long are those co-ordinates going to be valid for? It is a moving target isn't it and big scopes have small field of views (unless you are specifically seeking the use of a survey scope - but you didn't ask for one)

Cheers

[Joe Keller]

(I've posted additional information about this to Dr. Van Flandern's messageboard over the last few weeks.)

Uranus is +5.7 at opposition and almost 20 AU out. Barbarossa is almost 200 AU out. The brightness goes as the fourth power of distance because the inverse square law holds both ways (it's farther away, and also the sun is dimmer there). Four powers of 10, is 10 magnitudes. So, if Barbarossa were just like Uranus, it would be about +15.7 (more precisely, +15.9).

Uranus has a rather high albedo, 66% according to a 1983 college textbook. Albedos of 4 to 8% often have been used in recent years as "canonical" albedos in academic journal articles about comet nuclei and Kuiper Belt Objects. Also, a recent article theoretically estimated the albedo of one type of borderline brown dwarf, as 1%! So, it would be fair to use 6.6%, i.e., about 1/10 the albedo of Uranus: +15.9+2.5=+18.4. Using 1% albedo would give +20.45.

The objects I've found and suspected of being Barbarossa, have comparison albedos (these are necessarily somewhat inaccurate) ranging from +17.3 to +20, but mainly +18 to +19. Theoretically, Barbarossa (assuming 1 to 10 Jupiter masses) should be the size of Jupiter only if Barbarossa has a H/He composition like the sun or Jupiter. The same article predicts that if a medium-weight brown dwarf isn't made of hydrogen (even if it's helium or oxygen), it has 1/2 to 1/3 the diameter of the hydrogen version. Of course, this is only theory. That's why observation is so important.

[higginsdj]

OK, so has this object burnt off all it's H/He/O or just formed without it? (ie our sun sucked up all the H/He/O and blew off the rest for this brown dwarf to use.) Or are you basing this all on theories resulting in ever diminishing possibilities.

The only real problem I have with this 'discovery' is the fact that the discovery is based on historical plates in which 'objects' have disappeared rather than on something that has actually been detected.

[CometGuy]

Basically, if you dig around the noise of these images you'll always find a set of points that is a match to any orbit you desire.

This is why closely spaced observations over a couple of successive nights showing consistent motion would be needed to gain confidence that the object exists. For something near the detection limits of the scope you would need 5-6 observations over say a period of a week before anyone (i.e CBAT) took any interest.

[higginsdj]

Hi Joe,

Please do not think that I (and possibly others) are 'attacking' your theories. I for one take it as a given that there are between 1 and > 100,000 undiscovered objects in the outer solar system. I do not think anyone is questioning your theories and calculations. For all I know they may take real genius to perform. What I, and others, do question is the very first step - the 'observational' data on which all your work is based.

You have 1 ~40 year old image and 1 ~20 year old image for which you have not provided any evidence to show that the 'disappearing' dot is real. You have not, for each image on which the dot exists, shown another image taken by the same scope, same night to show that the dot was real and not something transient. The 3rd image that you published shows, to experienced observers, an image defect but you have chosen to ignore this advise and use your own experience.

Now the issue for me is that you have taken what amounts to flawed data, applied a lot of work (and I am not questioning any of the theory in that work) to produce a result that you want others to fork out lots of time and or money to observe. Flawed data in means flawed data out no matter how good/valid the work in between was. If you started off with quality data then you would have an army of amateur and professionals trying to find this object.

As a final note I would like to point out that one of the base requirements of good science is Peer review. The review may not agree with your findings. This does not make your results any more or less valid but taking the reviews that do not agree with your position personally and attacking them is just a waste of of your time - I don't care if you like them or not. Its up to the wider community to judge if good science has been practiced. The reviews themselves can be reviewed by others. Its not about right or wrong - it's about whether good practice has been overlooked, or a mistake has been overlooked or any number of other things that can affect a result. In other words, this is part of what is widely referred to as 'the scientific method'. A method developed and proven over a millennia......

Cheers

David