SpringerLink
Forum Springer Astron. Astrophys.
Forum Whats New Search Orders


Astron. Astrophys. 351, 619-626 (1999)

Previous Section Next Section Title Page Table of Contents

4. Luminosities and colours

Photometric measurements of the combined light of the system have been made by a number of authors over a broad range of colours. They are summarised and homogenized in the critical compilation of Leggett (1992), from which we adopt: U=10.79, B=9.57,V=8.09, R=7.09, I=5.97, J=4.75, H=4.14, K=3.93, L=3.77, L'=3.67, with quoted errors of [FORMULA]3% on IJHK. The system has colours consistent with those of solar metallicity stars (Leggett 1992), as expected from the spectroscopic metallicity discussed below.

The difference in brightness between the components of Gl 570BC has been measured on several occasions in different bandpasses. Mariotti et al. (1990) obtained magnitude differences of [FORMULA]=1.27[FORMULA]0.12, [FORMULA]=1.19[FORMULA]0.12, [FORMULA]=1.12[FORMULA]0.07, and [FORMULA]=1.06[FORMULA]0.17 from their 1D scanning speckle observations. HMcC published another estimate in the J band, [FORMULA]=1.30[FORMULA]0.04. Our own adaptive optics measurements provide [FORMULA]=1.18[FORMULA]0.04, as well as [FORMULA]=1.18[FORMULA]0.04 in a narrow band [FORMULA] (2.166[FORMULA]) filter. We estimate a correction of [FORMULA]=0.04, using synthetic photometry generated from NextGen (Hauschild, Allard, & Baron 1999) model spectra of effective temperatures that bracket those apropriate for M dwarfs of the luminosities of the two components of Gl 570BC. This provides a second determination of the K band flux ratio, [FORMULA]=1.22[FORMULA]0.04.

While those measurements are mutually consistent for every filter, their run with wavelenghth is only marginally compatible with the known IR colours of M dwarfs (Leggett 1992). The [FORMULA] and (particularly) [FORMULA] colours of early and mid-M dwarfs define a remarkably flat plateau (Leggett 1992): when the spectral type of solar metallicity stars (apropriate for Gl 570BC) varies between M0V (MK[FORMULA]4.8) and M5.5V (MK[FORMULA]7.9), [FORMULA] only changes from 0.68 to 0.57, and [FORMULA] just from 0.85 to 0.87 (Leggett 1992) The two components of Gl 570BC respectively have MK[FORMULA]5.4 and MK[FORMULA]6.6 and their [FORMULA] and [FORMULA] colours are therefore expected to only differ by about -0.04 and +0.01. [FORMULA] should therefore be almost identical to [FORMULA], whereas the observations give [FORMULA]-[FORMULA]=0.14[FORMULA]0.04, and [FORMULA]-[FORMULA] should be -0.04, whereas the observations indicate [FORMULA]-[FORMULA]=+0.11[FORMULA]0.12. As the integrated colours of the system match the expected values, this inconsistency must come from some of the measured magnitude differences. We suspect that it traces back to an overestimated contrast in the J and H band observations of both Mariotti et al. (1990) and HMcC, since speckle techniques have a known bias towards underestimating the relative flux of faint components (Perrier 1988). The K band adaptive optics observations are in principle immune to this bias, and the K band speckle observations of Mariotti et al. (1990) should be relatively unaffected, thanks to the low D/[FORMULA] (where D is the telescope diameter and [FORMULA] is the Fried parameter of the atmosphere) at this longer wavelength. We therefore tentatively adopt as the basis of our magnitude difference system the mean of the three K band measurements, [FORMULA]=1.18[FORMULA]0.03. From this value we then derive preferred values of [FORMULA]=1.19 and [FORMULA]=1.15, but we will also consider the published J and K flux ratios as an alternative. This discrepancy contributes significant uncertainty to the analysis, and better measurements of the flux ratios at J and H would be of considerable interest. To date all our adaptive optics measurements in the J and H bands have unfortunately been obtained at phases when the secondary star overlaps the first Airy ring of the primary for these wavelengths. These circumstances maximize the uncertainties in differential photometry from partially corrected adaptive optics images (Veran et al. 1999), and these data therefore contribute no useful information on the flux density ratio.

Absolute magnitudes were derived from the parallax and the apparent magnitudes of the individual stars. The absolute magnitudes of the brighter star have uncertainties which are dominated by those of the photometry of the system, while those for the secondary have some contribution from the magnitude difference. The parallax doesn't appreciably contribute in either case.


[TABLE]

Table 4. Absolute magnitudes of the two components


Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 1999

Online publication: November 3, 1999
helpdesk.link@springer.de