We present a model that predicts the light curve amplitude distribution for an ensemble of low-mass magnetically active stars , under the assumptions that stellar spin axes are randomly orientated and that cool starspots have a characteristic scale length and are randomly distributed across the stellar surfaces .
The model is compared with observational data for highly magnetically active M-dwarfs in the young cluster NGC 2516 .
We find that the best fitting starspot scale length is not constrained by these data alone , but requires assumptions about the overall starspot filling factor and starspot temperature .
Assuming a spot coverage fraction of 0.4 \pm 0.1 and a starspot to unspotted photosphere temperature ratio of 0.7 \pm 0.05 , as suggested by the inflated radii of these stars compared to evolutionary model predictions and by TiO band measurements on other active cool stars of earlier spectral type , the best-fitting starspot angular scale length is 3.5 ^ { +2 } _ { -1 } degrees , or a linear scale length of \sim 25 000 km .
This linear scale length is similar to large sunspot groups , but 2–5 times smaller than the starspots recently deduced on an active G-dwarf using eclipse mapping by a transiting exoplanet .
However , the best-fitting spot scale length in the NGC 2516 M-dwarfs increases with the assumed spot temperature ratio and with the inverse square root of the assumed spot filling factor .
Hence the light curve amplitude distribution might equally well be described by these larger spot scale lengths if the spot filling factors are < 0.1 or the spot temperature ratio is > 0.9 .