Traditional methods of identifying active galactic nuclei (AGNs) rely on UV and optical color criteria, but these methods exclude AGNs that are highly optically obscured by surrounding dust. This bias causes underestimates in the number of obscured AGNs in the Universe, and is problematic for our understanding of galaxy evolution and computing luminosity functions. Since much of the radiation absorbed by dust is re-radiated in the IR, mid-IR color selection provides a much more robust method of identifying AGNs that is not biased against highly dust obscured galaxies. The Wide-field Infrared Survey Explorer (WISE) has provided an all-sky mid-IR survey in four passbands, and the photometry from this catalog can be used to identify AGN candidates. We select AGNs from the catalog using a WISE $W1-W2>0.8$ color cut, and perform follow-up observations with the Keck DEIMOS optical spectrograph for redshift and emission line measurements. We spectroscopically confirm that mid-IR color selection does indeed identify obscured AGNs that are missed by the largest existing optical surveys, and targeted observations are necessary to find these objects. Combining WISE data with the AKARI far-IR all-sky catalog, we also perform SED analysis to characterize the physical properties of dusty AGNs and their surrounding obscuring medium. Using only photometry, we are able to extract a number of physical properties (ex. dust temperature and masses) without the need for time and resource-intensive spectroscopy. Finally, we test the effects of using different SED and color criteria to select different populations of AGNs from the 3D-HST catalog and compute their luminosity functions. We find that SED fitting can identify a large number of AGNs that are missed by traditional color selection methods, and this can potentially increase the sample size of AGNs available for analysis by several factors.

Ignoring redshift dependence, the statistical performance of a weak lensing survey is set by two numbers: the effective shape noise of the sources, which includes the intrinsic ellipticity dispersion and the measurement noise, and the density of sources that are useful for weak lensing measurements. We provide some general guidance for weak lensing shear measurements from a “generic” space telescope using the galaxy flux signal-to-noise ratio (S/N), ellipticity measurement variance, and effective galaxy number per square degree as metrics to evaluate weak lensing measurement performance across wavelength. Galaxy data are collected from both the ULTRAVISTA field and the Hubble Space Telescope (HST) Frontier Parallel Fields (FPFs) and fitted to radially symmetric Sersic galaxy light profiles. The Sersic galaxy profiles are then stretched to impose an artificial weak lensing shear, and then convolved with a pure Airy Disk PSF to simulate imaging of weak gravitationally lensed galaxies from a hypothetical diffraction-limited space telescope. Our results consistently show that weak lensing measurements are optimized for the longest wavelengths available in the survey, until thermal background from the telescope or interplanetary dust reduces the sensitivity. We also perform a crude measurement of galaxy ellipticities in the HST FPFs using Source Extractor (SExtractor) software (Bertin & Arnouts, 1996) in order to estimate a reduced shear at the location of the parallel fields. The average ellipticity among a significantly large density of distant galaxy images is directly related to reduced gravitational shear signal, which we approximate with our simplifying assumptions and use to estimate the primary FF galaxy cluster masses to illustrate that shear measurements at long NIR wavelengths are completely reasonable.

The measurement of the the Extragalactic Background Light (EBL) has seen some controversy in recent works, with direct and indirect measures conflicting. Specifi- cally, upper limits based on analyzing the plausible opacity obscuring TeV spectra of blazars suggests that the density of radiation with wavelengths near 3.4 μm is onethirdtoonehalfasintenseasdirectmeasuresofthesame(forexample: Aharonian et al., 2006; Levenson et al., 2007; Matsumoto et al., 2005). The dominant contributor of the EBL at 3.4μm is expected to be ordinary starlight from relatively local, z < 1, galaxies, so an estimate of the amount of light emitted by galaxies based on the galaxy Luminosity Function (LF) should provide a useful lower limit to the EBL. While analyses of this sort have been done by others (Dom ́ınguez et al., 2011; Helgason et al., 2012), the full sky coverage of the All- WISE database has made it possible for us to improve the measurement of both the LF at 2.4 μm and the EBL using the large public spectroscopic redshift surveys. In order to do so, we had to develop a mathematical model for the measurement of a generalization of the LF, which is the density of galaxies per unit comoving volume per unit luminosity, to the Spectro-Luminosity Functional (SLF), which replaces the density per unit single luminosity, dL, with the density per luminosi-

ii

ties at all frequencies, DLν. Our best combined analysis of the data yields present day Shechter Function LF parameters of: L⋆ = 6.4�[0.1stat, 0.3sys]�1010 L2.4μm ⊙ (M⋆ = −21.67�[0.02stat, 0.05sys] AB mag), φ⋆ = 5.8�[0.3stat, 0.3sys]�10−3 Mpc−3, and α = −1.050 � [0.004stat, 0.03sys]; this implies a present day density of galax- ies of 0.08 Mpc−3 brighter that 106 L2.4 μm ⊙ (10−3 Mpc−3 brighter than L⋆) and a luminosity density equivalent to 3.8 � 108 L2.4 μm ⊙ Mpc−3. The net EBL at 3.4μm that our synthesis model produces from galaxies closer than z = 5 is Iν = 9.0 � 0.5 kJy sr−1 (νIν = 8.0 � 0.4 nW m−2 sr−1), largely in agreement with similar LF based estimates of the EBL.