The roughness of a sheet of sea ice encodes its deformational history and determines its aerodynamic coupling with the overlying air and underlying water. Here we report snow surface, ice surface, and ice underside roughness computed from 47 surface elevation profiles collected during a transect of the Weddell Sea. The roughness for each surface, parameterized as the standard deviation of the surface elevation, segregates according to whether or not a floe has been deformed: deformed ice has greater roughness than undeformed ice. Regardless of deformational history, the underside roughness is almost always greater than the snow surface and ice surface roughnesses, which are nearly equal. Roughness spectra for all three surfaces and for both deformed and undeformed ice roll off roughly as k −1 when the wavenumber k is between 0.1 and 3 rad m −1 . The snow surface and underside spectra roll off somewhat faster than k −1 , and the ice surface spectra roll off somewhat slower than k −1 . Both top and underside Arctic ice roughness spectra, on the other hand, have been reported to roll off faster than k −2 . We speculate that the excess spectral intensity at high wavenumbers in the Antarctic ice surface spectra results from the small‐scale roughness that the ice sheet had on consolidation. This excess high‐wavenumber spectral intensity persists in the ice surface spectra of second‐year ice. Evidently, once formed, the ice surface remains unchanged on the microscale until the entire ice sheet melts. With a remote measurement of roughness, we should be able to decide whether an ice floe is deformed or undeformed. Our spectral analysis hints that remote sensing may also be able to differentiate between first‐year and second‐year ice. From the snow surface spectra, we compute a roughness scale ξ that parameterizes the air‐ice momentum coupling and lets us estimate the neutral stability drag coefficient referenced to a height of 10 m, C DN10 . Typical C DN10 values are 1.1–1.4 × 10 −3 over undeformed ice and 1.3–1.8 × 10 −3 over deformed ice.
No takes yet. Share an insight, caveat, or question.
Andreas et al. (1993) studied this question.