We present an analysis of X-ray high quality grating spectra of the Seyfert 1 galaxy NGC 5548 using archival Chandra HETGS and LETGS observations for a total exposure time of 800ks. The continuum emission is well represented by a powerlaw plus a blac
k-body component. We find that the well known X-ray warm absorber in this source consists of two different outflow velocity systems. Recognizing the presence of these kinematically distinct components allows each system to be fitted independently, each with two absorption components with different ionization levels. The high velocity system consists of a component with temperature of 2.7X10^6K and another component with temperature of 5.8X10^5K. The low-velocity system required also two absorbing components, one with temperature of 5.8X10^5K; the other with lower temperature (3.5X10^4K). Once these components are considered, the data do not require any further absorbers. In particular, a model consisting of a continuous radial range of ionization structures is not required. The two absorbing components in each velocity system are in pressure equilibrium with each other. This suggests that each velocity system consists of a multi-phase medium. This is the first time that different outflow velocity systems have been modelled independently in the X-ray band for this source. The kinematic components and column densities found from the X-rays are in agreement with the main kinematic components found in the UV absorber. This supports the idea that the UV and X-ray absorbing gas is part of the same phenomenon. NGC 5548 can now be seen to fit in a pattern established for other warm absorbers: 2 or 3 discrete phases in pressure equilibrium. There are no remaining cases of a well studied warm absorber in which a model consisting of a multi-phase medium is not viable.
(Abridged) We present a two month Suzaku X-ray monitoring of the Seyfert 1 galaxy NGC 5548. The campaign consists of 7 observations. We analyze the response in the opacity of the gas that forms the ionized absorber to ionizing flux variations. Despit
e variations by a factor of 4 in the impinging continuum, the soft X-ray spectra of the source show little spectral variations, suggesting no response from the ionized absorber. A detailed time modeling confirms the lack of opacity variations for an absorbing component with high ionization. Instead, the models tentatively suggest that the ionization parameter of a low ionization absorbing component might be changing with the ionizing flux, as expected for gas in photoionization equilibrium. Using the lack of variations, we set an upper limit of n_e <2.0E7 cm-3 for the electron density of the gas forming the high ionization, high velocity component. This implies a large distance from the continuum source (R > 0.033 pc). If the variations in the low ionization component are real, they imply n_e >9.8E4 cm-3 and R < 3 pc. We discuss our results in terms of two different scenarios: a large scale outflow originating in the inner parts of the accretion disk, or a thermally driven wind originating much farther out. Given the large distance of the wind, the implied mass outflow rate is also large (Mw > 0.08 Maccr). The associated total kinetic energy deployed by the wind in the host galaxy (>1.2E56 erg) can be enough to disrupt the interstellar medium, possibly regulating large scale star formation. The total mass and energy ejected by the wind is still lower than the one required for cosmic feedback, even when extrapolated to quasar luminosities. Such feedback would require that we are observing the wind before it is fully accelerated.
