The origin of the excess transit absorption in the HD 189733 system: planet or star? (1607.03684v2)

Abstract: We have detected excess absorption in the emission cores of Ca II H & K during transits of HD 189733b for the first time. Using observations of three transits we investigate the origin of the absorption, which is also seen in H{\alpha} and the Na I D lines. Applying differential spectrophotometry methods to the Ca II H and Ca II K lines combined, using respective passband widths of $\Delta\lambda$ = 0.4 & 0.6 \AA\ yields excess absorption of t$_d$ = 0.0074 $\pm$ 0.0044 (1.7$\sigma$; Transit 1) and 0.0214 +/- 0.0022 (9.8$\sigma$; Transit 2). Similarly, we detect excess H{\alpha} absorption in a passband of width $\Delta\lambda$ = 0.7 \AA, with t$_d$ = 0.0084 $\pm$ 0.0016 (5.2$\sigma$) and 0.0121 $\pm$ 0.0012 (9.9$\sigma$). For both lines, Transit 2 is thus significantly deeper. Combining all three transits for the Na I D lines yields excess absorption of t$_d$ = 0.0041 $\pm$ 0.0006 (6.5$\sigma$). By considering the time series observations of each line, we find that the excess apparent absorption is best recovered in the stellar reference frame. These findings lead us to postulate that the main contribution to the excess transit absorption in the differential light curves arises because the normalising continuum bands form in the photosphere, whereas the line cores contain a chromospheric component. We can not rule out that part of the excess absorption signature arises from the planetary atmosphere, but we present evidence which casts doubt on recent claims to have detected wind motions in the planet's atmosphere in these data.