Mid-transit times exhibit a sinusoidal variation with a
peak-to-peak amplitude of 1.5 +/- 0.7 minutes. If this is real it would
imply the presence of another exoplanet in the system, XO-1c. Since the
period of these variations is close to a year (350 days) no one will
believe this "signal" until it is confirmed by an independent analysis.
One-year "signals" in astronomy are notorious for being wrong! I can't
see how my JD to HJD for this object could be wrong when no other exoplanet
shows a one-year sinusoidal variation. I currently don't believe there's
evidence in this data set for supporting the presence of a real TTV.
Professional astronomer Burke et al (2010) have a preprint
(http://fr.arxiv.org/abs/1004.4252)
that doesn't find convincing evidence for a sinusoidal variation.
Basic data - Updated 2009.08.16
RA = 16:02:11.6, Decl = +28:10:11
Season = May 26
B = 11.85 ±
0.025, V = 11.19 ± 0.035, Rc = 10.81 ± 0.030, Ic = 10.43
± 0.040, B-V = 0.66 ±
0.05 (B. Gary, all-sky)
HJDo = 3808.9170 & P = 3.941534 day (discovery paper)
HJDo = 3887.74679 & P = 3.941534 (27) day (Schneider listing in Extrasolar Planets Encyclopaedia,
based on Wilson et al, 2006)
HJDo = 3887.7474 & P = 3.9415038
(72) day (AXA sinusoidal data fit, below, v20110312)
Depth = 23.21 ± 0.15 mmag (mostly BVRIC-band)
Length = 2.910 ±
0.015 hr (mostly BVRIC-band)
Fp (fraction of time in partial transit) = 0.26 ±
0.01
F2 (relative depth at contact #2) = 0.80 ±
0.02
Plot of all mid-transit measurements,
amateur and professional, versus expected HJD using the
ephemeris given in the upper-left had corner. The sinusoidal
model has been fit to the data using chi-squared minimization. The
peak-to-peak amplitude is 1.54 +/- 0.72 minutes (based on
chi-square procedures). This is a 2.1-sigma "detection" -
meaning that it probably is not significant.
This TTV plot is suggestive of a second planet in a resonant orbit that perturbs XO-1b's orbital motion. I am reluctant to interpret this data as constituting a "second planet" detection because my "eyeball" doesn't believe it. The amplitude is 3.6 times its formal SE. This illustrates what a search for another exoplanet would involve. In fact, the main message of this plot is that amateur data must be in groups of 6 to10 at the approximate same date in order for the median value of the group to be "useful." Every group has outliers, and the outliers have undue influence when they are not accompanied by other measurements at about the same date to reveal that they are outliers. It's too bad that observations in 2007 weren't coordinated with this in mind. If only groups of data are considered there are just 3 TTV values in this plot (one in 2006 and two in 2008) that can be relied upon for a TTV anomaly search. Another "problem" with the sinusoid solution in this plot is that the preiodicity is very close to one year. I'm confident that the JD to HJD conversion was done correctly, nevertheless, it is always prudent to be wary of any anomaly with a one year periodicity. In 2009 the sinusoidal model will be put to a definitive test I hope. (Symbol descriptions are available from the webmaster.)
Transit Parameters - Updated 2009.08.16
100622-xo1-C11(Ic)-pro
Note the correlated "brightening" above the model fit at 5.5
UT for simultaneously observed LCs using two telescopes.
Clouds ruined this observing session.
Many images had XO-1 saturated so the depth is
uncertain and the data are noisy.
0404GBL Depth
is lower than the others because that's the case for i'-band compared
with shorter wavelengths. Some autoguiding problems made for noisy data
early on.
8611roeV
8603gary
8603futR
8530futR
8413haar
8405gary High air
mass ruined the early data and clouds affected the
post-egress data.
8328gary
8109gary
7921gary Wind, clouds and
lightning degraded data & forced early closure.
6601futV
6601futR
6424heal Junk Bond
Observatory 32-inch allowed B-band observing. Depth
was greater, as expected.
6406vanR
6406vanV Depth
sortof deep. "Low frequency" variations.
6314gary This LC was used
to establish HJDo for the discovery paper ephemeris.
6306gary 2006.03.06.
UT mid ~12.88 (i.e., 11.40 + 1.48).
5712vanC
5705gary 2005.07.05.
UT mid ~03.74 +/- 0.10.
5701gary There were
lots of problems with this data: clouds, poor tracking,too-short
exposure times, etc.
More coming
6520holR
(this data downloaded from NStED)
Out-of-Transit (OOT) Light
Curves
In spite
of cirrus clouds, with losses up to ~1/2 magnitude,
the LC is featureless at the 1 mmag level.
I have a
couple dozen OOT observing sessions from May, 2006
that I'll add to this section when time permits.
Estimated SEs are 0.025, 0.035, 0.030, 0.040 magnitude,
based on RMS residuals of fit to three
Landolt star regions (N = 29, 35, 18, 14).
Note: The star to the upper-right of star #3 in the above
image is continuosuly variable ("contact binary"). Don't use it for reference.
RA/DE = 16:02:07.7,+28:14:45 (GSC 2041:1416). Two other stars to the
upper-right of this FOV are also variables (see next image).
Warning: Stars #3, #10 and #14 are variables.
Detailed description of Spring, 2006 observations of XO-1
McCullough et al, 2006 link
Wilson et al, 2006 link
Machalek et al, 2008 link
Return to calling web page AXA
____________________________________________________________________
