Constraining the history of the Sagittarius dwarf galaxy using observations of its tidal debris

We present a comparison of semianalytic models of the phase-space structure of tidal debris with measurements of average distances, velocities, and su rface densities of stars associated with the Sagittarius dwarf galaxy, comp iled from all observations reported since its discovery in 1994. We find th at several interesting features in the data can be explained by these model s. The properties of stars about +/- 10 degrees-15 degrees away from the ce nter of Sgr-in particular, the orientation of material perpendicular to Sgr 's orbit and the kink in the velocity gradient-are consistent with those ex pected for unbound material stripped during the most recent pericentric pas sage similar to 50 Myr ago. The break in the slope of the surface density s een by Mateo, Olszewski, & Morrison at b similar to -35 degrees can be unde rstood as marking the end of this material. However, the detections beyond this point are unlikely to represent debris in a trailing streamer, torn fr om Sgr during the immediately preceding passage similar to 0.7 Gyr ago, as the surface density of this streamer would be too low compared with observa tions in these regions. The low-b detections are more plausibly explained b y a leading streamer of material that was lost more that 1 Gyr ago and has wrapped all the way around the Galaxy to intercept the line of sight. The d istance and velocity measurements at b = -40 degrees reported by Majewski e t al. in a companion paper also support this hypothesis. We determine debris models with these properties on orbits that are consist ent with the currently known positions and velocities of Sgr in Galactic po tentials with halo components that have circular velocities v(circ) = 140-2 00 km s(-1). In all cases, the orbits oscillate between similar to 12 and s imilar to 40 kpc from the Galactic center with radial time periods of 0.55- 0.75 Gyr. The best match to the data is obtained in models where Sgr curren tly has a mass of similar to 10(9) M-. and has orbited the Galaxy for at le ast the last I Gyr, during which time it has reduced its mass by a factor o f 2-3, or luminosity by an amount equivalent to similar to 10% of the total luminosity of the Galactic halo. These numbers suggest that Sgr is rapidly disrupting and unlikely to survive beyond a few more pericentric passages. These conclusions are only tentative, because they rely heavily on the les s certain measurements of debris properties far from the center of Sgr. How ever, they demonstrate the immense potential for using debris to determine Sgr's dynamical history in great detail.