Although the character of the integrated spectrum of a galaxy
changes radically over the range of the Hubble sequence,
a revealing result in Fig. 1 is the subtlety of the changes in
composite spectrum between Hubble types E and Sb.
Most of the variance in spectral properties,
especially in the blue region, occurs among Sc-Irr galaxies.
The only exception is the 
emission feature,
which is sensitive to stellar population over the entire range of
types. This statement is quantified in
Fig. 6 by the clear correlation between EW (
)
and galaxy morphological type.
Therefore, it is not surprising that emission-line
luminosity and equivalent width are the most widely adopted star
formation tracers.
The 
luminosity provides a direct measure
of the global photoionization rate, which can be used in turn to
reliably estimate the SFR in massive (
) stars
by means of the stellar evolution and photoionization models
previously mentioned.
Figure 6: Distribution of 
emission equivalent width,
binned by RSA Hubble type. The symbol S denotes a Seyfert galaxy.
Extrapolation to a total SFR, integrated over all stellar masses, is
accomplished by adopting a given IMF.
Many IMFs can be approximated as power laws of the form
. In this notation the Salpeter
(1955) IMF is a single power law with 
= 2.35, while
the Miller & Scalo (1979) function can be fitted by 
=1.4
(0.1-1 
), 2.5 (1-10 
), and 2.3
(10-100 
).
The ``extended'' Miller-Scalo IMF adopted by
Kennicutt (1983,1994) is similar to the
Salpeter IMF above 1 
, but takes into account
the rollover in IMF at lower masses observed in the solar neighborhood:
With this IMF the relationship between the star formation rate and the
luminosity of a galaxy reduces to a single constant:
To derive with equations 5-6 a correct SFR
we must consider several factors that can affect
the 
flux measured (see the Praktikum section):
- the contamination of the 
emission by nonthermal
nuclear emission
This is normally not relevant
because nonthermal nuclear emission is negligible
(
in most of normal spirals.
- the underlying stellar 
absorption
absorption in the underlying red continuum should be
small as well since the light at those wavelengths is dominated by G-K
giant stars with a typical absorption EW of 1-2 Å.
- the 
emission in the 
profile
Average corrections for the larger effects of 
emission
are normally applied taking into account that the 
ratio is fairly constant in spiral and irregular galaxies,
spanning the range 0.75-0.95.
In this Praktikum we analyze high
resolution spectra (5-7 Å) of galaxies and we not need to apply
this correction because both lines are well separated.
Only for the Seyfert 2 galaxy we must take into account the
contamination from 
and perform a gaussian decomposition of the 
line profile.
- absorption by dust.
Extinction is by far the most important source of systematic
uncertainty in the SFR determination, whether measured from
or from modeling the broad-band colors.
We will discuss this point in detail in another section.
