Summary form only given. Ice-sheet models are necessary to understand changes being observed in Greenland and Antarctica, and predict ice sheets' response in a warming climate. Information on ice thickness, bed topography and basal conditions is required to accurately model ice dynamics. Most of the rapid changes are taking place around ice-sheets' margins with fast-flowing outlet glaciers. These areas are very rough and contain warm ice with debris and other inclusions, and thus low frequency radars are required to minimize volume scatter. Also these radars must have a 2-D aperture synthesis capability to reduce surface scatter from extremely rough surface because the surface and volume scatter can mask weak echoes from the ice-bed interface. Therefore, a UAS that can fly on a closed-spaced lines in the cross-track direction for synthesizing a 2-D aperture is needed. In view of this, we have developed a compact UAS, G1X with 5.3 m wingspan, equipped with a 2-kg radar. The onboard radar operates at 14 MHz and 35 MHz with bandwidths of about 1 MHz and 4 MHz, respectively. The G1X carries two separate antennas integrated conformally onto its wings. Because of the small platform, it was extremely challenging to integrate the long-wavelength antennas as well as the onboard metallic mechanical structures, electronics boxes and wirings. Managing electromagnetic interference between avionics and radar systems also requires close collaboration in aircraft system and payload system design in a small UAS. In our presentation, we will first describe our radar system and platform and then we will focus on the design and implementation of the HF/VHF antennas for UAS integration. During the UAS operation in the field, we have characterized the radar antennas inflight and data collected were used to optimize the antenna bandwidths and thus the radar data quality. Radar echograms collected before and after the antenna optimization will be presented for comparison. Finally, we will show ice-sheet thickness data collected during remote-controlled and autonomous flights as part of the science mission near the Whillans ice stream in West Antarctica during the Winter 2013 field season.

We developed a compact radar for use on a small UAV to conduct measurements over the ice sheets in Greenland and Antarctica. It operates at center frequencies of 14 and 35 MHz with bandwidths of 1 MHz and 4 MHz, respectively. The radar weighs about 2 kgs and is housed in a box with dimensions of 20.3 cm x 15.2 cm x 13.2 cm. It transmits a signal power of 100 W at a pulse repletion frequency of 10 kHz and requires average power of about 20 W. The antennas for operating the radar are integrated into the wings and airframe of a small UAV with a wingspan of 5.3 m. We selected the frequencies of 14 and 35 MHz based on previous successful soundings of temperate ice in Alaska with a 12.5 MHz impulse radar [Arcone, 2002] and temperate glaciers in Patagonia with a 30 MHz monocycle radar [Blindow et al., 2012]. We developed the radar-equipped UAV to perform surveys over a 2-D grid, which allows us to synthesize a large two-dimensional aperture and obtain fine resolution in both the alongand cross-track directions. Low-frequency, high-sensitivity radars with 2-D aperture synthesis capability are needed to overcome the surface and volume scatter that masks weak echoes from the ice-bed interface of fast-flowing glaciers. We collected data with the radar-equipped UAV on sub-glacial ice near Lake Whillans at both 14 and 35 MHz. We acquired data to evaluate the concept of 2-D aperture synthesis and successfully demonstrated the first successful sounding of ice with a radar on an UAV. We are planning to build multiple radar-equipped UAVs for collecting fine-resolution data near the grounding lines of fast-flowing glaciers.

One necessary parameter for flight dynamic analysis of any unmanned aircraft is its moment of inertia. Moment of inertia is an inherently difficult and time consuming parameter to measure. This paper presents several ways to calculate moment of inertia of unmanned aerial vehicles. In addition, it also presents the development and application of a correlation that estimates moment of inertia of a particularly configured aircraft. The correlation is designed such that only easily measured parameters are needed to estimate the moment of inertia. It is found that this correlation method mostly underestimates the moment of inertia.