Monday, January 14, 2013

Jerusalem WS lecture notes: 11. The physics of stellar feedback

By M. Krumholz, the slides.
  • 'conservative weed'
  • Bate 2009 simulation -- SC formation. SF too efficient and fast -- SF efficiency close to 100%
  • what inhibits SF? feedback:
  • hot gas or photons push material away from the star, kinetic energy in the material shell = star energy output | radius set by momentum conservation (energy or momentum driven cases, radiation or winds). Mass in the shell way larger than the wind mass
  • feedback budgets:
    • Q -- radiant energy, wind energy, number of ionising photons
    • IMF-averaged production rate: luminosity per unit mass (~M/L ratio)
    • lifetime-weighted production rate -- energy out of unit mass (e.g. ergs/g)
    • stochastic IMF sampling in dwarfs -- SLUG code
    • galactic wind: at least as much mass as went into stars
    • what feedbacks are interesting? those that can cause velocities higher than escape velocities --> lower limt
    • losses: gravity, collisions (loss of momentum)
  • ISM feedback taxonomy:
    • ionising radiation: not important for galactic winds formation (sound speed ~10 km/s, so can influence in smaller MCs, Krumholz 2006, 2009, Dale 2012), probably the most important SF regulator today
    • radiation pressure (photon momentum, Thompson scattering) -- ~200 km/s -- cannot be responsible for galactic winds, unless radiation enhancing fraction f_{trap} >> 1. Can be important for subgalactic objects, dwarf galaxies, can blow up gas clouds.
    • the important question: what is the f_{trap}?:
    • 30 Dor: dust grain temperature can help infer the IR radiation field, Lopez 2011
    • simulations: 2D, high resoluton: RT instability -- similar to oil floating on water, right panel: no gravity, 2 different optical depths[surface densities]: RP may affect sub-galactic objects, but cannot produce galactic winds, Krumholz & Thompson 2013
    • stellar winds: Solar wind is a wimpy old thing, O stars. Momentum driven
    • 30 Dor -- most massive binary star system, each ~83 M_{\odot}, still on the MS
    • supernovae: energy budget in stars of 8-10 M_{\odot}
      • N_{SN}/M = 0.01 M_{odot}
      • less energy and momentum than radiation feedback
      • more energy, less momentum than winds
      • SN are most important because they are much closer to energy conserving feedback -- large velocities, post-schock ejecta temperatures are ~10^{10} K --> cooling time is ~ 60 Myr, whereas time required to escape the galaxy is << 1 Myr, so gas cannot cool
      • Sedov-Taylor similarity solution -- first developed for nuclear tests, open literature only in 1995 --> energy of Trinity blast from Time pictures (R_blast as a fn of time, Sedov),
      • trapping factor ~30-40, momentum goes up by this factor during the energy conserving phase, density dependent: SNe explode in low density environments due to star radiation --> f_{trap} is elevated
      • SNs can dominate momentum budget --> proper simulation should take other feedbacks into accounts
    • metallicity feedback
    • metallicity changes SF law (makes difference in dwarfs, high z galaxies) --> metallicity regulated SF (Kuhlen 2012 simulation), interactions wih other feedbacks

No comments:

Post a Comment