While galaxies are star-forming, the telltale signature of their spectra are strong emission lines. Emission lines are an extremely valuable "tool". The strength and shape of the various lines allow to constrain physical properties. Further, they provide a unique signature for save redshift determinations even at lowest masses and highest redshifts. Finally, emission lines can be used in the first place to select large samples of galaxies at well defined redshifts through narrowband (NB) surveys, as the presence of a line in the NB filter raises the flux compared to that expected from the continuum.

In principle, a direct product from NB surveys are line fluxes for the selected objects. However, as the effective bandpasses of NB interference filters in the fast convergent beams of survey telescopes cannot be made with tophat bandpass shapes, flux measurements for individual objects have uncertainties. We applied a novel NB measurement method to a sample of UltraVISTA NB118 H α emitters, and demonstrate that this technique allows to recover more accurate fluxes. Additionally, the method allows to determine more accurate redshifts.

Contemporary NB surveys have a strong focus on searches for Ly α emitters at the highest redshifts. Ly α is so well suited for high redshift studies because it is the intrinsically brightest emission line and has a low rest-frame wavelength. However, as it is a resonant line, it can be diluted through scattering at neutral hydrogen. Consequently, it can be be used as a probe for the gas surrounding a galaxy.

Around some objects the extended Ly αemission is so strong that it can be detected for individual objects. In this thesis extremely deep VLT/XSHOOTER rest-frame far-UV spectroscopy is presented for Himiko, a gigantic Ly α emitter at redshift z = 6.6 or a time when the universe was only 6 per cent of its present age. We constrained with our work, which of the possible Ly α producing mechanisms is responsible for the majority of Himiko’s Ly α emission. We conclude that normal Pop II star formation is the most likely explanation.

Himiko is a rare object. Yet, recent studies have shown that weak extended Ly α emission might be a typical property around star-forming galaxies at high redshifts; but a recent non-detection of such emission for a sample of Ly α emitter z ¡­ 2 has raised the question whether the halo properties change with time. As this more typical Ly α emission is to weak to be detected for individual objects, stacking analysis is required. We performed such an analysis with an independent sample and find a halo consistent with measurements at higher redshifts.