... star TWA 5

Appendix A is only available in electronic form at http://www.edpsciences.org

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

... Sun

Only four TWA members have measured Hipparcos distances, whose average value, 55 pc, has been assumed as the distance of TWA 5.

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

...

Table A.1 is available at the CDS, and it contains the following information for each observed line: observed and predicted line wavelength (Cols. 2 and 3), element and ionization state (Col. 4), electronic configurations of the atomic levels (Col. 5), temperature of maximum emissivity (Col. 6), observed line flux (Col. 7).

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

... TWA 5

We note that the MCMC method preferentially explores ${\it EMD}$ bins which are best constrained by the selected emission lines. Since error estimation depends on the quality of the sampling, statistical uncertainties are estimated only for those ${\it EMD}$ bins explored many times (Kashyap & Drake 1998).

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

... model

This is not obvious since our RGS ${\it EMD}$ model was obtained from analysis of selected line fluxes and not from a global spectral fitting.

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

... sample

However, the results for L1551 region are based on low resolution EPIC spectra, and therefore may not be directly compared to ours.

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