Groundbased radar is a key technique for the post-discovery reconnaissance of NEOs and is likely to play a central role in identification of possibly threatening objects during the foreseeable future. Delay-Doppler measurements are orthogonal to optical angle measurements and typically have a fractional precision between 10^-5 and 10^- 9, and consequently are invaluable for refining orbits and prediction ephemerides. The same measurements can provide two-dimensional images with resolution on the order of decameters. Imaging data sets with adequate coverage in subradar longitude/latitude can be used to determine the target's shape and spin vector. The active planetary radars use wavelengths that are sensitive to near-surface bulk density and structural scales larger than a few centimeters and, for comets, can penetrate optically opaque comas and reveal large-particle clouds. Upgrades of existing telescopes (especially Arecibo) will expand the range of groundbased radar and will optimize NEO imaging and astrometric capabilities. However, existing instruments are already oversubscribed, and observation of more than a small fraction of objects discovered in a Spaceguard-like survey will require radar telescopes dedicated to NEO reconnaissance.