Understanding HST Observations of IGM and ISM Clouds

This proposal requests support to extend and enhance the plasma code Cloudy to better simulate ntergalactic and interstellar clouds. ISM and IGM gas is far from equilibrium and best understoo by reference to complete numerical simulations (Osterbrock & Ferland 2006; hereafter AGN ). These simulations are coupled with HST spectroscopic observations to determine for t e chemical evolution and star formation history of the Universe. Hydro simula ions can do an excellent job of reproducing many large-scale structures. But they must comprom se on the microphysics, often using simple fitting functions, due to the complexity. It is quite co on to see Cloudy used either as the sole aid (for example, Aracil et a1.2006), or a supplement to ynamics calculations (for example, Scannapieco et a1.2006), in interpreting HST IGM and ISM bsorption lines. Cloudy, like all microphysics codes, must compromise other aspects of the s mulations to make them possible on today's computers (Ferland & Savin 2001). These include e assumption that atomic processes are time steady, that the gas is static, relatively simple geomet ies, and an approximate treatment of radiative transfer. The proposed extensions will greatly im rove the representation of complex geometries that occur in the ISM and IGM. The plasma c de Cloudy was developed to do a complete simulation of the microphysics that occurs within non-equilibrium gas and predict the emitted spectrum. The code's abilities have always been Ii ited by the speed of available computers. The highest priority has been a complete treat ent of the microphysics that occurs within the gas. The code has been applied to a wide range 0f henomena because of this attention to the microphysics. If all processes that govern how at ms emit or absorb light, and are ionized and recombine, are included, then that simulation can e applied to gas in any environment. This goal has been largely realized. The code goes to th ISM and strict thermodynamic equilibrium limits, and to the fully ionized and fully molecula limits. Simulations ave always required, and always will require, some form of simplified treatment to make the calcu ations tractable. As machines grow ever faster the fidelity of the simulation can be improved. Clo dy was designed for the case of illumination by a star, a point source. These clouds are actu lly illumina':ed by the cosmic background or a diffuse starburst. This project will upgrade the Ii and continuum radiative transport to fully capture the lJhysics of the interactions. The geometry nd radiative transport methods, currently ID spherical and escape probabilities, will be extend d to exact methods for an arbitrary geometry. Cloudy is op nly available on the web (http://wvv\\'.nublado.org). Figure 1 shows that over 100 papers per yea now use Cloudy in some way, with roughly a quarter of these involving HST observations. he full list of papers is given on the Cloudy web site. A cloud illu inated by backgtound radiation The geometry now assumed in most plasma codes is show in the left panel of Figure 2. Light strikes a cloud as a ray from a 90 degree angle (by default). he first step in the project is to incorporate an explicit integration over solid angle so that radiatia from an arbitrary distribution of angles can be considered. The case of isotropic illumination, a propriate for the IG\1background and shown in the right panel, can be done analytically in losed form. Nurrerical quadrature will be used for greater flexibility. There are tw effects of illumination by an isotropic radiation field, appropriate for ionization by the backgroun ,compa:-ed with single-ray iLuminatioll, appropriate for a point source like a star or quasar. Light ntering the surface at a shallow angle will be absorbed closer to the surface due to the increased ath bngth.