This white paper emphasizes the potential of QSO absorption lines in the rest-frame far/extreme UV at rest-frame wavelengths from ~500 to 2000 A. In this wavelength range, species such as Ne VIII, Na IX, and Mg X can be detected, providing diagnostic
s of gas with temperatures >> 10^{6} K, as well as banks of adjacent ions such as O I, O II, O III, O IV, O V, and O VI (and similarly N I - N V; S II - S VI; Ne II - Ne VIII, etc.), which constrain physical conditions with unprecedented precision. A UV spectrograph with good sensitivity down to observed wavelengths of 1000 A can detect these new probes in absorption systems with redshift z(abs) > 0.3, and at these redshifts, the detailed relationships between the absorbers and nearby galaxies and large-scale environment can be studied from the ground. By observing QSOs at z = 1.0 - 1.5, HST has started to exploit extreme-UV QSO absorption lines, but HST can only reach a small number of these targets. A future, more sensitive UV spectrograph could open up this new discovery space.
Using high-resolution UV spectra of 16 low-z QSOs, we study the physical conditions and statistics of O VI absorption in the IGM at z < 0.5. We identify 51 intervening (z_{abs} << z_{QSO}) O VI systems comprised of 77 individual components, and we fi
nd 14 proximate systems (z_{abs} ~ z_{QSO}) containing 34 components. For intervening systems [components] with rest-frame equivalent width W_{r} > 30 mA, the number of O VI absorbers per unit redshift dN/dz = 15.6(+2.9/-2.4) [21.0(+3.2/-2.8)], and this decreases to dN/dz = 0.9(+1.0/-0.5) [0.3(+0.7/-0.3)] for W_{r} > 300 mA. The number per redshift increases steeply as z_{abs} approaches z_{QSO}, and some proximate absorbers have substantially lower H I/O VI ratios. The lower proximate ratios could be partially due to ionization effects but also require higher metallicities. We find that 37% of the intervening O VI absorbers have velocity centroids that are well-aligned with corresponding H I absorption. If the O VI and the H I trace the same gas, the relatively small differences in line widths imply the absorbers are cool with T < 10^{5} K. Most of these well-aligned absorbers have the characteristics of metal-enriched photoionized gas. However, the O VI in the apparently simple and cold systems could be associated with a hot phase with T ~ 10^{5.5} K if the metallicity is high enough to cause the associated broad Ly alpha absorption to be too weak to detect. We show that 53% of the intervening O VI systems are complex multiphase absorbers that can accommodate both lower metallicity collisionally-ionized gas with T > 10^{5} K and cold photoionzed gas.
