Polaris the Cepheid returns: 4.5 years of monitoring from ground and space

The Astrophysical Journal 683, 433−440, 2008

Polaris the Cepheid returns: 4.5 years of monitoring from ground and space

H. Bruntt
School of Physics, University of Sydney, NSW, Australia

N.R. Evans
Smithsonian Astrophysical Observatory, Cambridge, MA, USA

D. Stello
School of Physics, University of Sydney, NSW, Australia

A.J. Penny
University of St Andrews, School of Physics and Astronomy, St. Andrews, UK

J.A. Eaton
Center of Excellence in Information Systems, Tennessee State Univ., Nashville, TN, USA

D.L. Buzasi
Dept. of Physics, US Air Force Academy, Colorado Springs, CO, USA

D.D. Sasselov
Astronomy Dept., Harvard University, Cambridge, MA, USA

H.L. Preston
Dept. of Physics, US Air Force Academy, Colorado Springs, CO, USA

E. Miller-Ricci
Astronomy Dept., Harvard University, Cambridge, MA, USA

Abstract

We present the analysis of 4.5 years of nearly continuous observations of the classical Cepheid Polaris, which comprise the most precise data available for this star. We have made spectroscopic measurements from ground and photometric measurements from the WIRE star tracker and the SMEI instrument on the Coriolis satellite. Measurements of the amplitude of the dominant oscillation (P = 4 days), that go back more than a century, show a decrease from 120 mmag to 30 mmag (V magnitude) around the turn of the millennium. It has been speculated that the reason for the decrease in amplitude is the evolution of Polaris towards the edge of the instability strip. However, our new data reveal an increase in the amplitude by about 30% from 2003-2006. It now appears that the amplitude change is cyclic rather than monotonic, and most likely the result of a pulsation phenomenon. In addition, previous radial velocity campaigns have claimed the detection of long-period variation in Polaris (P > 40 days). Our radial velocity data are more precise than previous datasets, and we find no evidence for additional variation for periods in the range 3-50 days with an upper limit of 100 m/s. However, in the WIRE data we find evidence of variation on time-scales of 2-6 days, which we interpret as being due to granulation.