1: Dear Dr. Sneden,
2:
3: We have reviewed the referee's helpful comments and made most of the changes he/she suggests. Our responses to his/her individual comments are below prefaced with "***". Please let me know if something is unclear.
4:
5: Thank you,
6:
7: Evgenya Shkolnik
8:
9: *********************Referee's Report*************************************
10:
11: In the extended version of the paper, the authors present new results regarding a very interesting system, a SB4. I like this version very much. The results are now better constrained and are really interesting.
12:
13: However, I would suggest the authors to go a little bit further and add few things, which might help the scope of the paper and the understanding by the readers.
14:
15: One interesting result is the well defined orbit of the eclipsing system A (Aa+Ab) and the implications on different fields. The separation is 0.035 AU. Since the radii of each component is 0.61 Rsun, the separation is few times this value. They are K7, so they have important convective envelopes. These facts mean that both circularization and synchronization processes are very effective. They enhance activity due to fast rotation is also very important. So, my recommendation is that the authors discusses these processes in a separate subsection. For the tidal effect on circularization and synchronization, they should read:
16: - Zahn 1977 (already mentioned in the text) and 1989
17: - Tassoul 1987, 1988
18: - Goldman & Mazeh 1991
19:
20: *** The referee makes an interesting suggestion to compare the results of our tight, eclipsing binary with the existing theories on tidal circularization and synchronization processes. Each of the three paper listed above conclude that the circularization timescale is proportional to the orbital period to some power. The powers published are 16/3, 49/12 and 10/3 for Zahn, Tassoul and Goldman & Mazeh, respectively. The circularization timescale for this system according to all three theories is <= 70 Myr. Our system would only help distinguish between these theories if a) it had a period closer to the observed "transition" period of circularization of 5.7 days or if we had a precise age determined for the system that was less than 70 Myr. Unfortunately, neither of these are the case, in fact the system is most certainly older than that. Therefore binary Aa and Ab of BD -225866 cannot distinguish between these three published theories.
21:
22: This binary (Aa+Ab) can be classified as a BY Dra system. See Strassmeier et al. 1993 (A catalog of chromospherically active binary stars, second edition). How different and similar is this system to others included in this catalog?
23:
24: *** We have compared the Aa+Ab pair with the six late-K and early-M systems in the Strassmeier et al. 1993 catalog, and indeed they certainly do show common characteristics such as: Ca II, UV, X-ray, H_alpha emission. We have included this in the text.
25:
26: Actually, this system seems to be a perfect target for Doppler imaging. So I suggest the authors to contact experts in this field for the follow-up.
27:
28: *** The Doppler imaging possibility is interesting but maybe not practical given the complex braiding of the various spectroscopic components. v sin i for the A pair is high enough to give some surface resolution, but not a lot. Zeeman Doppler Imaging (ZDI) might be more interesting. At V=10 it's certainly doable with CFHT's spectropolarimeter, ESPADONS, though there are other late-K single stars and close pairs for which we could get a cleaner result more easily.
29:
30:
31: One easy analysis that I miss is the age estimate based on the derived masses and radii. Since the system seems to be a BY Dra, they can compare with the results by Barrado et al. 1994. In any case, any age estimate for the Aa+Ab eclipsing binary will be extended to the quadruple system and could be included in the discussion about the orbital configuration and the need of the disk (sections 4 and 5). by the way, a HR diagram with the location of the components (specially Aa and Ab) would be nice.
32:
33: *** The mass-age relationship from Barrado et al. 1994 is not tested for stars much less massive than the Sun as is the case for the K7+K7 binary pair (each 0.588 Msun). If we use the published relationship, the resulting age is 37 Gyr (!), only confirming our conclusion that the system in not young. Unfortunately, actually stating an age at this point is not possible. We can only say the system is "old", >120 Myr, the age of the Pleiades.
34: An HR diagram with Aa&Ab does not convey any new information. The use of photometric distances means the components are guaranteed to lie in the right place.
35:
36:
37: Regarding this issue (the age estimate), the authors discuss it in page 7 (section 3), using activity and lithium as proxies. However, since the close binary is tidally locked, activity CANNOT be used as age indicator (activity is enhance in close binaries). In the same way, a shown by Deliyannis et al. (1994) in subgiants belonging to M67, Barrado et al. (1996) in dwarf binaries belonging to the Hyades, and Barrado et al. (1997) in chromospherically active binaries, lithium is artificially preserve in tidally locked binaries. Therefore, it cannot be used as an age estimate. It is quite curious that the authors cannot detect lithium. The 21 mA upper limit is much smaller than measurements by Soderblom et al. (1993) for the Pleiades (120 Myr) and equivalent to the few datapoints in the case of the Hyades (about 700 Myr).
38:
39: *** We certainly agree with the referee that we cannot use activity as an age indicator for a tidally locked system. In Section 3 we state that the enhanced activity is due to the synchronous rotation of the two stars rather than youth. If the system were young, (<70 Myr) then we should also see emission features from the second pair (Ba and Bb) which we do not. The lack of lithium in the spectrum only confirms that the system is not young.
40:
41:
42: Regarding the periodogram (figure 4), can they comment on the other two possible periods, at 0.7 and 1.8 days?
43:
44: *** We had a closer look at the two periods in the periodogram. One was at 1.8220910 days, the other was at 0.6864408 days. When phase-folding the data, both periods produce phase-folded light curves that, in general, look like a sine-curve but not nearly as well phased as with the 2.21 period of the strongest peak. Although we cannot absolutely rule out other stellar periods, there is no reason we would adopt a lesser peak in the periodogram to be the rotation period of the star. In addition, the short 0.68 day period is not consistent with the vsini width of the spectral lines. For a radius of 0.61Rsun, the 0.68 day period would produce a vsini~ 45 km/s which is certainly not the case given the width of the spectral lines.
45: Also, the strongest peak at P_rot=2.21 days agrees with the orbital period of the close binary, consistent with the fact that the system is old and therefore expected to be tidally locked.
46:
47:
48: Minor points:
49:
50: Section 5.
51: " ... within 2000 AU of the system." I wold say that no member down to magnitude xxx in the filter yyy has been found, which translates to a maximum mass of zzzz Msun using whatever distance and age.
52:
53: *** Good point. We have added a statement setting these limits: "We searched the Naval Observatory Merged Astrometric Dataset (NOMAD; Zacharias et al.~2005) which compiles data from several catalogs including the 2MASS catalog (Skrutskie et al. 2006). With a K-band magnitude limit of 16, $M_K$ is limited to 12.4 mags at the system distance of 51 pc. This sets the maximum mass of any potential companion to 20--50 M$_J$, the mass of an old field mid- to late-L dwarf."
54:
55: Figure 5. The fit is barely visible.
56:
57: *** We have fixed this.
58:
59: Figure 7. I would add the theoretical models for age= 50, 500 and 5000 Myr (for example) to illustrate the effect of age and
60: the evolution on the relationship.
61:
62: *** We have added the models of these ages which reinforce the fact that the system is old.
63:
64:
65: