My research is centered on the resolved stellar populations of the Local Group (LG), and in particular on Galactic globular clusters (GGCs) and dwarf galaxies. For these stellar systems, color-magnitude diagrams (CMDs) can be obtained, from which the information about their past star formation history (SFH) can be extracted. In particular, dwarf galaxies of the LG show large variations in their mass, environmental conditions, current SF, gas content and morphology, so one hopes that possible correlations of their SFHs with these parameters will shed light on the evolutionary mechanisms of these relatively simple stellar systems, and possibly of their larger ``normal'' counterparts (elliptical and spiral galaxies).
This rationale motivated, as a first step, the investigation of three ``special''
dwarfs, namely the Tucana dwarf spheroidal (dSph), the Phoenix dwarf spheroidal/dwarf
irregular (dSph/dIrr), and the SagDIG dIrr galaxy (Saviane I., Held
E. V., Piotto G., 1996, A&A, 315, 40; Held E.V., Saviane I., Momany
Y., 1999, A&A, 345, 747; Momany Y., Saviane I., Held E.V., 2000, in preparation).
Almost all LG dSph are located near the two giant spirals, but Tucana and Phoenix
are an exception, being 
kpc far from either the Milky Way or M31.
In turn, SagDIG is the LG dwarf with one of the highest gas/stars ratio (
).
Moreover, all three galaxies contain (or were suspected to contain) young/blue
stars, so it was expected that imaging done with the new generation, larger
format and blue-sensitive CCDs, would have provided precious information.
New instrumentation, and properly chosen observational strategies, are able to uncover subtle evolutionary details even for ``well studied'' objects. This is demonstrated by our studies of Fornax and Leo I. The CMD that was obtained for the Fornax dwarf spheroidal galaxy (Saviane, Held & Bertelli, 2000, A&A, 355, 56), allowed us to study a number of previously undetected features. The improvement is due to the increased area coverage that was offered by the DFOSC imager attached to the ESO/Danish telescope. Even better coverage is provided by the new WF imager attached to the ESO/MPI 2.2m telescope, and indeed a medium-term observational program was started at ESO and is now being carried out.
In the case of Leo I, we have been able, for the first time, to unambiguously reveal the presence of a true old population (Held, Saviane, Momany & Carraro, 2000, ApJ, 530, L85). Our success, in comparison to even extensive HST studies, was due to the choice of bluer colors and to the spatial coverage of the outer regions of the galaxy.
Even the study of simple stellar systems requires basic input such as distance indicators, metallicity indices, and reliable stellar evolutionary models. Models can be tested using simple stellar populations (SSPs) such as globular clusters. However, internal dynamical processes likely affect the evolution of their stars. Then I decided to study a peculiar GGC (NGC 1851) in order to see if theoretical predictions can be reconciled with the observed CMDs (Saviane I., et al., 1998, A&A, 333, 479). Moreover, the need to obtain the metallicity distribution in old stellar systems, led me to an extensive analysis of the behavior of the morphology of the red giant branch (RGB) as a function of the metallicity (Saviane et al. 2000, A&A, 355, 966). The study was based on the largest homogeneous GGC photometric database, that was assembled for the relative age investigation (see below).
How does the SFH of LG dwarfs compare to that of other local galaxies? Is there a common zero age population? In order to begin answering these questions, a long-term observational campaign was planned in 1996, aimed at collecting homogeneous photometric data for a large sample of inner-halo Galactic globulars. The first phase of this project is now complete, and a thorough analysis of 34 CMDs is now yielding a better view of the first stages of the SF in our Milky Way (Rosenberg, Saviane, Piotto, Aparicio, 1999, AJ, 118, 2306). A complementary study of young (or suspected young) GGCs is also under way. This should clarify later stages of GC formation (e.g. a second SF episode, a delayed GC formation in outer halo, later accretion events, etc.). Two objects (Palomar 1 and Palomar 12) were studied up to now (Rosenberg A., Saviane I., Piotto G., Held E.V., 1998, A&A, 339, 61; Rosenberg A., Piotto G., Saviane I., Aparicio A., Gratton R., 1998, AJ, 115, 658; Rosenberg A., Saviane I., Piotto G., Aparicio A., Zaggia S. R., 1998,AJ, 115, 648). The first quantitative evaluation of their ages has been given, confirming the young Pal 12 age, and leading to the discovery of the youngest Galactic globular, Pal 1.
The best summarizing statement of all the previous points, is that stars in
the LG began forming at the very same time throughout a (current) volume of
Mpc. Later, it is almost impossible to find common trends in SF.
However, we have hints that two major SF episodes were experienced by many dwarfs,
and that the second episode occurred in more central regions. This could also
be true for the Milky Way halo, but a check with data for more peculiar GCs
is needed.