Statistics – Computation
Scientific paper
May 2009
adsabs.harvard.edu/cgi-bin/nph-data_query?bibcode=2009aas...21441117c&link_type=abstract
American Astronomical Society, AAS Meeting #214, #411.17; Bulletin of the American Astronomical Society, Vol. 41, p.678
Statistics
Computation
Scientific paper
Astrometry alone does not constrain the direction of circulation of the star in its reflex motion orbit. Practically, this means that astrometry cannot distinguish between the ascending and descending nodes in the reflex motion orbit. Consequently, Omega (the longitude of ascending node) and omega (the longitude of periastron, which is measured with respect to Omega in the direction of the orbit) can only be determined modulo 180 degrees. Radial velocity data can provide the information needed to break this degeneracy.
When combining RV and astrometric data to fit an orbit, one must check that the fitted orbit is physically consistent with both data sets. The astrometric orbit problem can be partially linearized via transformation to four Thiele-Innes parameters plus three nonlinear parameters: eccentricity, period and periastron time. On the other hand, the RV orbit problem can be partially linearized via transformation to two additional Thiele-Innes parameters plus the same three nonlinear parameters. Unfortunately, the two RV Thiele-Innes parameters are not linearly related to the four astrometric ones. Because of this difficulty, currently available algorithms for jointly fitting RV and astrometric data to multiple-planet systems employ five nonlinear parameters per planet. We have developed an algorithm that satisfies the RV-astrometry consistency requirement, while using only three nonlinear parameters per planet. We have successfully tested this algorithm in the context of a recent double-blind planet detection simulation study.
We expect the reduction in nonlinearity to give the algorithm a significant advantage in computation speed over existing algorithms, making it ideally suited to carry out further simulation studies of joint astrometric and RV orbit-fitting for multiple-planet systems via application of computation-intensive Bayesian methods based on Markov Chain Monte Carlo.
This work was carried out at the Jet Propulsion Laboratory, California Institute of Technology, under contract with NASA.
No associations
LandOfFree
Multiple-planet Orbit-fitting with Joint Radial Velocity and Astrometric Data does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.
If you have personal experience with Multiple-planet Orbit-fitting with Joint Radial Velocity and Astrometric Data, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Multiple-planet Orbit-fitting with Joint Radial Velocity and Astrometric Data will most certainly appreciate the feedback.
Profile ID: LFWR-SCP-O-1106838