Astron. Astrophys. 327, 1114-1122 (1997)

4. Conclusions

UBV photoelectric photometry aimed at studying flare activity on EV Lac has lead to the detection of a total of 254 flares in 1272 h of monitoring time.

We find that three percent of the energy emitted by EV Lac in the U -band is due to flare emission, while this energy fraction in the B and V bands goes down to 0.3% and 0.06%, respectively. A comparison with previous results in the literature shows no significant variation of the flare energy distributions with time.

The flare activity level computed on a yearly time-scale shows a slight variability with a period of 3 year beginning from 1970. This result is similar to Mavridis & Avgoloupis's (1986) result. Their data on flare occurrence, from 1971 to 1980, show the same time variability in the years 1971-1976 as our data. It should be noticed, however, that these authors suggest a period of flare activity close to 5 years.

Flare energies in U, B and V bandpasses were found to be strongly correlated; the slope of the - correlation is consistent with the slope derived by Lacy et al. (1976) from analysing flares on different flare stars.

The observed values of the flare rise time and decay time are slightly correlated to the flare energies. The correlation is better and steeper for the decay time than for the rise time. Flares of equal amount of energy output can be characterized by 1-2 order of magnitude different time-scales. A comparison of the empirical relation between the rise time and the energy in the U-band found by us and that one derived by Pettersen (1989) from a wide sample of flare stars, yields to the conclusion that the relation that holds for EV Lac is definitely steeper than the general one; i.e. EV Lac flares with a certain value of energy are characterized by a mean rise-time higher than the value predicted by the general law given by Pettersen (1989).

Seasonal light curves from 1969 to 1972 reveal that EV Lac exhibited rotational modulation only in 1971.

We have investigated the behaviour of the flare occurence rate versus the rotational phase. Our data allowed us to ascertain a well defined rotational modulation of the flare occurence rate in 1970. There is a strong spatial correlation between the site of high flare activity level in 1970 and the site in which spots were clustering in the successive observational season. We suggest that this scenario is reminiscent of the solar one: flares occur more frequently in region of emerging magnetic flux, i.e. where spots are growing up (Rust 1972, van Hoven et al. 1980, Priest et al. 1986). This result is in agreement with the conclusion by Doyle (1987) that now can be more strictly constrained by our data.

© European Southern Observatory (ESO) 1997

Online publication: April 6, 1998
helpdesk.link@springer.de