Jm. Brett, OPACITY SAMPLING MODEL PHOTOSPHERES FOR M-DWARFS .1. COMPUTATIONS, SENSITIVITIES AND COMPARISONS, Astronomy and astrophysics, 295(3), 1995, pp. 736-754
We have computed a new grid of model photospheres for M dwarfs which c
overs the parameter ranges 4000 K less-than-or-equal-to T(eff) less-th
an-or-equal-to 2600 K, log g = 4.5, 5.0 and abundances of (1.0, 2.0, 0
.1, 0.01) times the solar value. Although these models retain the stan
dard assumptions of 'classical models' they are computed (for the firs
t time) with the accurate and realistic opacity sampling technique. Th
is has been applied to an extensive set of atomic and molecular line o
pacites computed with laboratory values of transition strengths, which
will permit a critical comparison to observations. We outline the com
putation of the models and describe the sensitivity of the computed st
ructures to the input parameters T(eff), log g and abundance as well a
s the mixing length parameter l/h. Interesting effects connected with
molecular thermodynamics are found in the coolest models. Comparisons
to other model photospheres for M dwarfs demonstrate the importance of
having sufficient opacity sources, good opacity technique and correct
solar abundances for the resulting model structure. Despite the much
improved treatment of opacity, a comparison to observations reveals pr
oblems which are highlighted in the coolest models by the strength and
sensitivity of H2O absorption bands. An analysis of this problem indi
cates that it probably due to the models being too cool in the upper p
hotospheric layers. Two likely causes are identified as the neglect of
chromospheric heating and the simple treatment of convection employed
. More sophisticated models will be required to investigate this probl
em further.