Search for p-mode oscillations in DA white dwarfs with VLT-ULTRACAM. I. Upper limits to the p-modes

TL;DR

This study used high-resolution VLT-ULTRACAM data to search for p-mode oscillations in DA white dwarfs near the blue edge of the DAV instability strip, but found no definitive detections and set very low upper limits on amplitude.

Contribution

First high-resolution search for p-modes in DA white dwarfs near the blue edge, establishing the lowest amplitude upper limits to date.

Findings

No p-modes detected in any targets.

Upper limits on amplitude are below 0.1%.

Possible low-amplitude power excess observed in G 185-32.

Abstract

The main goal of this project is to search for p-mode oscillations in a selected sample of DA white dwarfs near the blue edge of the DAV (g-mode) instability strip, where the p-modes should be excited following theoretical models. A set of high quality time-series data on nine targets has been obtained in 3 photometric bands (Sloan u', g', r') using ULTRACAM at the VLT with a typical time resolution of a few tens of ms. Such high resolution is required because theory predicts very short periods, of the order of a second, for the p-modes in white dwarfs. The data have been analyzed using Fourier transform and correlation analysis methods. Results: P-modes have not been detected in any of our targets. The upper limits obtained for the pulsation amplitude, typically less than 0.1%, are the smallest limits reported in the literature. The Nyquist frequencies are large enough to fully cover the frequency range of interest for the p-modes. For the brightest target of our sample, G 185-32, a p-mode oscillation with a relative amplitude of 5x10**(-4) would have been easily detected, as shown by a simple simulation. For G 185-32 we note an excess of power below ~2 Hz in all the three nights of observation, which might be due in principle to tens of low-amplitude close modes. However, neither correlation analysis nor Fourier transform of the amplitude spectrum show significant results. We also checked the possibility that the p-modes have a very short lifetime, shorter than the observing runs, by dividing each run in several subsets and analyzing these subsets independently. The amplitude spectra show only a few peaks with S/N ratio higher than 4 sigma but the same peaks are not detected in different subsets, as we would expect, and we do not see any indication of frequency spacing. As a secondary result of this project, ... (see the paper)