Satellite and aircraft observations have revealed that remarkable changes in the Earth's polar ice cover have occurred in the last decade. The impacts of these changes, which include dramatic ice loss from ice sheets and rapid declines in Arctic sea ice, could be quite large in terms of sea level rise and global climate. NASA's Ice, Cloud and Land Elevation Satellite-2 (ICESat-2), currently planned for launch in 2015, is specifically intended to quantify the amount of change in ice sheets and sea ice and provide key insights into their behavior. It will achieve these objectives through the use of precise laser measurements of surface elevation, building on the groundbreaking capabilities of its predecessor, the Ice Cloud and Land Elevation Satellite (ICESat). In particular, ICESat-2 will measure the temporal and spatial character of ice sheet elevation change to enable assessment of ice sheet mass balance and examination of the underlying mechanisms that control it. The precision of ICESat-2's elevation measurement will also allow for accurate measurements of sea ice freeboard height, from which sea ice thickness and its temporal changes can be estimated. ICESat-2 will provide important information on other components of the Earth System as well, most notably large-scale vegetation biomass estimates through the measurement of vegetation canopy height. When combined with the original ICESat observations, ICESat-2 will provide ice change measurements across more than a 15-year time span. Its significantly improved laser system will also provide observations with much greater spatial resolution, temporal resolution, and accuracy than has ever been possible before.

High radiometric resolution imagery from the Advanced Land Imager (ALI) can detect the spatial reflectance variations associated with sastrugi (snow dunes) on ice sheets. This is shown, in theory, by considering the appropriate SNR ratio of ALI as compared with eight-bit optical sensors and, in practice, with multitemporal ALI imagery. Comparison of ALI imagery spanning 7.5 months also shows that the reflectance pattern associated with sastrugi can persist over this interval offering the potential of their use for measuring ice sheet velocity in uncrevassed regions.

Two Hyperion images of the Greenland ice sheet are used to characterize errors in the visible near-infrared (VNIR) and shortwave infrared (SWIR) detector arrays of Hyperion. Spatial variability in detector output is seen in both arrays and, in both cases, is largest at the shorter wavelengths and decreases as wavelength increases. Standard deviations of the maximum variability are 40 digital numbers (DNs) for the VNIR array and 123 DNs for the SWIR array. Based on a single pair of images, temporal stability is nearly as large and exhibits the same spectral characteristics of large variability at shorter wavelengths and decreasing variability with increasing wavelength. The uniformity and stability of the ice-sheet surface enables a detailed characterization and emphasizes the utility of using ice sheets as targets to achieve on-orbit sensor characterization.

Sea level change occurs on all timescales, depending on the type of change in question. It also occurs with a continuous range of spatial scales-local, regional, and global. To understand and be able to eventually predict sea level changes is a truly interdisciplinary endeavor. It requires geodetic and non-geodetic measurements of various types from space as well as in situ, while various numerical models for a number of meteorological and geophysical processes or properties are essential or relevant.

A digital elevation model (DEM) produced from airborne interferometric synthetic aperture radar (TOPSAR) data of a 5 km/spl times/65 km region of the Greenland Ice Sheet is compared with multiple airborne laser altimeter profiles of the same surface. TOPSAR-derived elevation values were calculated on a 10 m/spl times/10 m grid. Surface topography at wavelengths greater than one kilometer is accurately reproduced. A planar adjustment of -8 m with along-track and cross-track tilts of 0.17 and 2.4 m/km, respectively, is required to match the two elevation data sets. Residual noise of the TOPSAR data at subkilometer wavelengths is less than two meters. This study underscores the need for independent elevation data for controlling TOPSAR elevation surveys.