Abstract
We examine the Ly-alpha absorber population at z<0.3 detected in spectra of the QSOs PG0953+415 and H1821+643 taken with the Space Telescope Imaging Spectrograph aboard the Hubble Space Telescope. We compare their statistical properties to those in carefully-constructed mock quasar spectra drawn from a cosmological hydrodynamic simulation of a Lambda-CDM universe. We find very good agreement in the column density and b-parameter distributions, down to the smallest observable absorbers with N_HI\approx 10^{12.3} cm^-2. The observed absorber population is complete for N_HI\ga 10^{13} cm^-2, with a column density distribution slope of \beta=2.04\pm 0.23 and a median b-parameter of 21 km/s above this limit. The intergalactic gas giving rise to these weak absorbers is analogous to that at high redshift, located in diffuse large-scale structures that are highly photoionized by the metagalactic UV flux, though a greater number arise within shock-heated warm gas. The density, temperature, and column density of these absorbers follow similar relationships to those at high redshift, though with substantially larger scatter due to the shock-heated gas. The b-parameters typically have a significant contribution from thermal broadening, which facilitates a measurement of the low-z IGM temperature as traced by Ly-alpha absorbers. From our simulation we estimate T_IGM\sim 5000 K, with an upper limit of 10^4 K, at the mean density. The agreement in predicted and observed amplitude of the column density distributions allows us to measure the HI photoionization rate at z=0.17 to be 10^{-13.3\pm 0.7} s^{-1} (estimated modeling uncertainty), close to predictions based on quasar properties.