... cloud

Appendix B is only available in electronic form at http://www.aanda.org

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...

Appendix A is only available in electronic form at the CDS via anonymous ftp to cdsarc.u-strasbg.fr (130.79.128.5) or via http://cdsweb.u-strasbg.fr/cgi-bin/qcat?J/A+A/468/379

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...1,

New address (since September 2006): Integral Science Data Centre, Ch. d'Ecogia 16, 1290 Versoix, Switzerland & Geneva Observatory, University of Geneva, Ch. des Maillettes 51, 1290 Sauverny, Switzerland;

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... curves

For point sources identified in both the small $0\hbox{$.\!\!^{\prime\prime}$ }5 \times 0\hbox{$.\!\!^{\prime\prime}$ }5$ and the larger $1\hbox{$^{\prime\prime}$ }\times 1\hbox{$^{\prime\prime}$ }$ central windows, we used count rates and magnitudes from the small window because of the smaller pixel bin size and of the more frequent time sampling.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... magnitude

$\log {{\rm Rate_{OM}}} = -0.4 (m - Z)$, where m is the source's magnitude, and Z is the zero-point magnitude.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... values

When more than one extinction magnitude was available for multiples, we used the value from the primary.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... respectively)

The OM light curves were obtained by manually determining the position of the two stars in each exposure image, and then using an IDL routine to extract the light curve. We used radii of half the stars' separation (3 $.\!\!^{\prime\prime}$48, i.e., a radius of 3.65 pixels for the central window), which minimizes the contamination of one source onto the second one. We corrected for aperture, applied the theoretical and empirical corrections, corrected for deadtime, and applied a time-dependent sensitivity correction. The above procedure is similar to what the SAS task omsource does; however, the latter cannot yet extract for radii lower than 6 pixels. We defined an annulus of radii 20 and 30 pixels ($\approx$ $10\hbox{$^{\prime\prime}$ }$ and $15\hbox{$^{\prime\prime}$ }$) for the local background contribution. See Grosso et al. (2007a) for a detailed description of the procedure.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... Tau

In the case of DN Tau, due to a feature in SAS 6.x, about half the counts were processed in the exposure S009 (DATE-OBS = "2005-03-05T01:37:35'' and DATE-END = "2005-03-05T02:02:35'') only. The average count rate in this exposure is 37.7 ct s^-1, after correction for deadtime, aperture, and coincidence losses. The detected count rate for S009 is 32.2 ct s^-1 (i.e., 1.17 times smaller), based on the theoretical correction in the XMM-Newton Users Handbook (the empirical correction is minimal). In contrast, the average count rate in the preceding exposure (S408) is 87.7 ct s^-1. The latter corresponds to a detected count rate of 63.4 ct s^-1 (i.e., 1.38 times smaller). Thus, we multiplied the count rates in S009 by $2 / 1.17 \times 1.38 = 2.36$ to correct for the SAS feature. For DI Tau, the pointing of the satellite changed unexpectedly after DATE-OBS = "2005-02-09T19:30:13'' (exposures S412, S009, S413-S416), and DI Tau fell just on the edge of the OM FAST window. We thus discarded count rates after this epoch. For BP Tau, a problem with the time stamps of exposure S051 occurred. The DATE-OBS and DATE-END keywords of the following exposure, S052, were incorrectly used, and thus the time stamps of events were incorrectly calculated by the SAS software. We manually corrected these time stamps and keywords by using the correct keywords, DATE-OBS = "2004-08-16T06:45:56'' DATE-END = "2004-08-16T08:27:51'', and incidentally shifting the time stamps by -6256 s. The same timing problem for the S051 IMAGING data of BP Tau needed a similar correction of the date keywords. 

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.