COMMISSION 27 OF THE I. A. U.
           INFORMATION BULLETIN ON VARIABLE STARS
                        Number 3303

                                      Konkoly Observatory
                                      Budapest
                                      14 March 1989

                                      HU ISSN 0374 - 0676


           A RELATION BETWEEN X-RAY SURFACE FLUXES
          AND U-BAND "TIME SIGNATURES" IN dMe STARS


Utilizing the MAC technique (Andrews 1988a) for analyses of photometric
data in the Johnson U-band, we now have "time signatures" for seven
dMe stars, which describe their brightness fluctuations in terms of rapid
quasi-periodic oscillations. We have previously suggested that these time
signatures (TS) are possibly related to the X-ray luminosities of the stars
(Andrews 1988b). There are usually several TS's per star in the range
investigated, from about 6 seconds to a few minutes, depending to some extent
on the sampling times (usually continuous 1-second integrations collected over
10 to 20 minutes). When several TS's are present, the lower TS's may be
harmonics, and, of course, are limited by the sampling Nyquist frequency,
so we have examined the relationship of the maximum TS detected for each
star with the X-ray luminosity. In Fig. 1, we give a plot of log L_x against
log (1/TS_max), the logarithm of the maximum frequency of oscillations (in
Hertz). We are presently looking for a connection with either non-radial
oscillations or a possible link with the Ionson-Mullan hypothesis that
convective motions in the presence of strong magnetic fields may contribute to
coronal heating (Ionson 1984, Mullan 1984). Temporal and spatial
coherence in a large coronal loop may allow detection of oscillations against
the low surface brightness of a dMe star. There are no predicted values for
asteroseismological p-mode oscillations in dMe stars. However, for comparison,
we include in the plot the observed 10-minute "p-mode feature" for
the K2 dwarf, epsilon Eridani, discovered in the variation of the strength of
the CaII H and K lines (Noyes et al. 1984). This departs from the 4-minute
asteroseismological period predicted for a K2 star by Christensen-Dalsgaard
and Frandsen (1983). A value of log Lx = 28.29 for epsilon Eridani (=
Gliese 144) is listed by Bookbinder (1985), and values for the dMe stars are
available from Bookbinder (1985) and Pallavicini (1988). N.B. There is an
error in log (Lx) quoted previously for V1054 Oph (Andrews 1988b), where
Gliese 644C was used instead of 644AB, and should be 29.08.


                          [FIGURE 1]
                          [FIGURE 2]


In the log Lx versus log (1/TS_max) diagram we find considerable scatter,
and V645 Cen (Gliese 551) which has a low X-ray luminosity destroys any
relation if it exists. We have therefore examined the possible relation with
the X-ray surface flux, F_x, (see Figure 2). V645 Cen (Prox Cen) is appreciably
shifted in relation to the other stars due to its relatively small radius
compared with those of the other dMe stars, but V645 Cen still stands out
and may possess a larger TS. In Figure 2, we show two linear fits obtained
by minimizing the square deviations in ordinate and abscissa. Excluding the
dK2 star and V645 Cen, an approximate mean relation is found which is
valid over log F_x=6.4 to 7.0 : log(1/TS_max)=1.75 log F_x-13.32,
where the time signature is in seconds, and the X-ray surface flux is in
ergs per sq.cm per second. For the six stars the correlation coefficient is
0.89, and we have the power law: frequency(Hz) ~ (Fx)^1.75, for the
U-band oscillations. We predict an X-ray surface flux of 6.0 for epsilon
Eridani compared with 5.64 deduced from Bookbinder's (1985) observations.
We urge flare star observers to examine especially their U-band monitoring
for low X-ray emitters. Amongst the dMe stars, EZ Aqr=Gliese 866
(log F_x=5.66), has a predicted modulation, TS_max, of 43 minutes
in the U-band. The dM1.5 star, AX Mic (Gliese 825) is a good southern
hemisphere candidate for a search due to its brightness (V=6.7 mag) and
very low X-ray flux (log L_x=27.28, log F_x=4.76), which is predicted
to modulate with an unlikely 27.1 hour period according to the above power
law. This may be explicable by the fact that AX Mic is not a dMe star
(no Balmer emission). We tentatively point out that the small-amplitude
photometric oscillations of 5 to 30 hours in some K-type Pleiades members
(van Leeuwen 1983) are possibly a phenomenon related to that suspected in
the solar neighbourhood dMe stars.


A.D. ANDREWS
Armagh Observatory,
Armagh BT61 9DG
N.Ireland

References

Andrews, A. D.: 1988a, IAU Comm. 27. Inf. Bull. Var. Stars. No. 3197. and
   Astron. Astroph. (in press)
				[BIBCODE 1989A&A...210..303A ]
Andrews, A. D.: 1988b, IAU Comm. 27. Inf. Bull. Var. Stars. No. 3248.
Bookbinder, J. A.: 1985, Ph.D. thesis, Harvard University. [BIBCODE 1985PhDT........13B ]
Christensen-Dalsgaard, J. and Frandsen, S.: 1983, Solar Physics. 82, 469. [BIBCODE 1983SoPh...82..469C ]
Ionson, J. A.: 1984, Astroph. Journ. 276, 357. [BIBCODE 1984ApJ...276..357I ]
Mullan, D. J.: 1984, Astroph. Journ. 282, 603. [BIBCODE 1984ApJ...282..603M ]
Noyes, R. W., Baliunas, S. L., Belserene, E., Duncan, D. K., Horne, J. and
   Widrow, L.: 1984, Astroph. Journ. 285, L23. [BIBCODE 1984ApJ...285L..23N ]
Pallavicini, R.: 1988, Activity in Cool Star Envelopes, eds. Havnes, O. et
   al., (Kluwer Acad. Publ.), 249. [BIBCODE 1988ASSL..143..249P ]
van Leeuwen, F.: 1983, Ph.D. thesis, Leiden. [BIBCODE 1983PhDT........33V ]