Supernovae: Obscured or Intrinsically Dark?: Supernova surveys, canonically carried out at optical wavelengths, appear to be missing a significant fraction of supernovae: the measured supernova rate is a factor of ~2 lower than expected from the star formation rate in the local Universe (Horiuchi et al. 2011). One way to suppress the supernova rate is if many massive stars end their lives in a ``dark" event (like direct collapse to a black hole) or explode with significantly lower energy than typically thought.
An alternative explanation for the discrepancy between supernova rate and star formation rate is that a large fraction of core-collapse supernovae are severely dust obscured and therefore missed by our optically-selected samples (Mattila et al. 2012). Radio observations are not subject to this bias, and indeed, radio observers have already stumbled upon several supernovae without optical counterparts (Gal-Yam et al. 2006, Brunthaler et al. 2009). How many supernovae might we be missing due to simple dust obscuration? By comparing the radio-selected supernovae with optical supernovae detected by Pan-STARRS1 and the Liverpool Telescope, CHILES VERDES will obtain the first direct measurement of the fraction of supernovae hidden by severe dust extinction.

The Merger Rate of Compact Objects (and their EM signatures): Mergers of compact objects in binary systems (either black holes and/or neutron stars) are some of the most promising sources for detectable gravitational radiation. While gravitational wave observatories like LIGO and Virgo will soon have the requisite sensitivity to detect such mergers out to hundreds of Mpc (Abadie et al. 2010), the results from these detectors will be difficult to interpret for two reasons. First, the rate of mergers in compact binaries is wildly uncertain (to three orders of magnitude). Secondly, gravitational wave observatories achieve only poor localization of detections, and therefore the site and physical cause of the gravitational radiation will remain largely unknown without the identification of an electromagnetic counterpart.
Merging neutron stars are expected to produce bright radio signals lasting for 5 years, upon expulsion of near-relativistic material (Nakar & Piran 2011, Piran et al. 2013). The radio counterpart to such mergers should be distinguishable from other astrophysical blast waves from the amplitude and duration of the light curve, radio spectral evolution, and multi-wavelength counterpart (or lack thereof; Piran et al. 2013). Any constraint on the neutron star merger rate is of significant interest, as it is a critical factor driving operations and detection strategies at gravitational wave observatories and a key uncertainty in differentiating the site of r-process nucleosynthesis (Qian 2012).

Rate and Energetics of Gamma-Ray Bursts: Gamma-ray bursts must be beamed towards us to be detected by high-energy facilities like Swift and Fermi, and we therefore miss ~99% of such energetic events. However, the afterglows of gamma-ray bursts are much more isotropic, and radio transient surveys should be able to detect orphan afterglows from off-axis long gamma-ray bursts in significant numbers. The detection of an orphan afterglow, or limits on the population of orphan afterglows, places independent constraints on the opening angle of gamma-ray burst jets, their energetics, and the true rate of long gamma-ray bursts (Rhoads 1997, Perna & Loeb 1998, Totani & Panaitescu 2002). A detection of an orphan afterglow would achieve one of the longest-sought goals of gamma-ray burst studies, while non-detection of orphan afterglows will place strong limits on the total rate and energy of gamma-ray bursts.

Launching of Relativistic Jets from Supermassive Black Holes: Tidal disruption events occur when a star wanders too close to a quiescent super-massive black hole, becomes disrupted, and is accreted; they may be some of the most common and luminous radio transients (Frail et al. 2012). Tidal disruption events are of particular interest because they can probe the demographics of black hole mass (and potentially spin) (e.g., Strubbe & Quataert 2009, Kesden 2012). Tidal disruption events were predicted to be bright radio transients by Giannios & Metzger (2011) and van Velzen et al. (2011), and this claim was almost immediately observationally confirmed by Swift J164449.3+573451, an unusually luminous event which produced an on-axis jet (Bloom et al. 2011, Zauderer et al. 2011).
If all tidal disruption events produce jets, even off-axis jets should become visible as radio transients ~1--10 year after disruption. With the current small samples, it remains unclear if all tidal disruption events produce jets at some point in their evolution, or if some tidal disruption events never produce jets at all (perhaps due to low black hole spin; Krolik & Piran 2012). Our deep radio transient survey will place independent limits on the fraction of tidal disruption events that produce jets, without requiring thermal emission as a trigger. This strategy side-steps a significant source of uncertainty, because the thermal emission from tidal disruption events may peak anywhere between the optical and X-ray bands, making it difficult to carry out a robust search for this thermal signature. Tidal disurption events can also be used as high-precision laboratories for the study of conditions leading to the formation of relativistic jets.

The Known Unknowns: Of the 12 robust radio transients identified in blind surveys, two still lack a clear physical explanation (Bower et al. 2007, Frail et al. 2012, Jaeger et al. 2012). They are only present in a single observing epoch and are not co-located with an optical host galaxy or any multi-wavelength counterpart. These transients could be explained as dust-obscured long GRBs viewed on-axis or short GRBs offset from the host galaxy (Frail et al. 2012). Alternatively, their properties are consistent with a giant flare from an extragalactic magnetar or a flare from a neutron star in the Milky Way (Ofek et al. 2010). The nature of these radio transients will be throughly explored with CHILES VERDES, as we expect to detect ~100 of these mysterious transient events, and they will be followed up contemporaneously with multi-wavelength data.

The Variable Radio Sky: A detailed understanding of the transient radio sky also requires an understanding of the variable radio sky. When a source appears above our detection threshold, has it suddenly ``turned on" or has it simply brightened? An accurate answer to this question requires a thorough characterization of the amplitudes and timescales of radio variability. With the depth and time coverage of CHILES VERDES, we will be able to measure exquisite light curves for hundreds of variable radio sources, and we will also determine if fainter radio sources are more variable as recently suggested by Hodge et al. (2013).