The Lyman-α forest absorption seen in the foreground of high-redshift quasars have been a crucial probe of the z>2 Universe for several decades. Quasars have been crucial for this task because they are luminous, with >105 observable even with relatively small telescopes like SDSS.
However, push into faint magnitudes (>23rd mag) with a large telescope (8-10m), and truly ubiquitous star-forming Lyman-Break galaxies (LBGs) start coming into view: at 24th mag, several hundred LBGs within a narrow redshift range (Δz~0.5) can be observed within 1 sq deg, with separations of ~2-3 arcmin which corresponds to 2-3 h-1Mpc transverse comoving distance. In other words, the foreground Lyman-α forest sightlines from an array of LBGs would be able probe this kind of scale, allowing Wiener-filtering (or other techniques) to reconstruct the full 3D absorption field, which traces the underlying cosmic web large-scale structure.
See Lee et al 2014a for a detailed discussion on the feasibility of IGM tomography on existing and near-future facilities.
CLAMATO aims to obtain spectra for ~1500 LBGs within ~1 sq deg area in the COSMOS field, in order to generate a 3D map with comoving dimensions of ~70h-1Mpc × 70h-1Mpc × 300h-1Mpc, equivalent to a volume of (100 h-1Mpc)3.