We use a sample of 17 strong gravitational lens systems from the BELLS GALLERY survey to quantify the amount of low-mass dark matter haloes within the lensing galaxies and along their lines of sight, and to constrain the properties of dark matter. Based on a detection criterion of 10σ, we report no significant detection in any of the lenses. Using the sensitivity function at the 10-σ level, we have calculated the predicted number of detectable cold dark matter (CDM) line-of-sight haloes to be µ l = 1.17 ± 1.08, in agreement with our null detection. Assuming a detection sensitivity that improved to the level implied by a 5-σ threshold, the expected number of detectable line-of-sight haloes rises to µ l = 9.0 ± 3.0. Whilst the current data find zero detections at this sensitivity level (which has a probability of P 5σ CDM (n det = 0)=0.0001 and would be in strong tension with the CDM framework), we find that such a low detection threshold leads to many spurious detections and non-detections and therefore the current lack of detections is unreliable and requires data with improved sensitivity. Combining this sample with a subsample of 11 SLACS lenses, we constrain the half-mode mass to be log(M hm ) < 12.26 at the 2-σ level. The latter is consistent with resonantly produced sterile neutrino masses m s < 0.8 keV at any value of the lepton asymmetry at the 2-σ level.

We present a study of seventeen LAEs at redshift 2< z <3 gravitationally lensed by massive early-type galaxies (ETGs) at a mean redshift of approximately 0.5. Using a fully Bayesian grid-based technique, we model the gravitational lens mass distributions with elliptical powerlaw profiles and reconstruct the UV-continuum surface brightness distributions of the background sources using pixellated source models. We find that the deflectors are close to, but not consistent with isothermal models in almost all cases, at the 2σ-level. We take advantage of the lensing magnification (typically µ 20) to characterise the physical and morphological properties of these LAE galaxies. From reconstructing the ultra-violet continuum emission, we find that the star-formation rates range from 0.3 to 8.5 M yr −1 and that the galaxies are typically composed of several compact and diffuse components, separated by 0.4 to 4 kpc. Moreover, they have peak star-formation rate intensities that range from 2.1 to 54.1 M yr −1 kpc −2 . These galaxies tend to be extended with major axis ranging from 0.2 to 1.8 kpc (median 561 pc), and with a median ellipticity of 0.49. This morphology is consistent with disk-like structures of star-formation for more than half of the sample. However, for at least two sources, we also find off-axis components that may be associated with mergers. Resolved kinematical information will be needed to confirm the disk-like nature and possible merger scenario for the LAEs in the sample.

Lyman-alpha emitting (LAE) galaxies are thought to be predominantly responsible for the re-ionisation of the Universe and are, as such, one of the most studied star-forming galaxy populations. Current optical and narrow-band studies are limited by the angular resolution of the observations and the considerable investment in telescope time. Strong gravitational lensing is an extremely powerful method that can be used to overcome these limitations. In my talk I will present a study on the first homogeneous sample of 17 lensed Lyman-alpha emitters at redshift 2 < z < 3. By taking advantage of the lensing magnification, I was able to access the detailed structure of this high redshift star-forming galaxies, finding that they have radii ranging from 0.2 to 1.8 kpc and have a complex and clumpy morphology, with a median ellipticity of 0.49. This is consistent with disk-like structures of star-formation, which would rule out models where the Lyman-alpha emission is only seen perpendicular to the disk, and favours those clumpy models for the escape lines of sight for Lyman-alpha photons. We also find that the star formation rates range from 0.3 to 8.5 Mȯ/yr and that these galaxies tend to be very compact. The lower limit to their intrinsic size is about a factor of two smaller than that found for non-lensed LAEs, which highlights the power of gravitational lensing and sophisticated lens modelling techniques for resolving such objects in the high redshift Universe.