Author(s): Mikko Lensu
Linked Author(s):
Keywords: No Keywords
Abstract: The realiable ice thickness determination is crucial for the parameterisation ice models. As thickness cannot be remotely sensed a vast effort is being put to the retrieval of ice freeboard by satellite altimeters (IceSat, CryoSat). The thickness is then estimable using density and snow data. Freeboard variat ion is described by a distribution. Distributions sampled during field campaigns usually refer to short scales (point determinations of freeboard) while those for the two altimeters refer to 70 or 300 m freeboard averages respectively. Between the two scale regimes there exists a certain connection. Understanding of the freeboard distribution and the scale connection is required for the algorithms that retrieve the freeboard from altimeter echos and, on the other hand, for the derivation of sub-footrprint variation from the altimeter data. A basic question is the determination of the relative amount of ridged ice. In short length scales, which resolve individual ridges, the freeboard distribution is a superposition that can be separated into a normally distributed level ice component and into a ridged ice component which is of exponential type. In longer length scales the freeboard values average over both ice types so that the ridging signal grows weaker. The scale connections relevant for the retrieval of ridging by satellite altimeters are studied from 2D ice surface profile scanned from the Bay of Bothnia. In addition to distribution separation the extremal statistics is relied upon. The extremal statistics is dominated by ridging features and the ridging signal is more persistent in longer length scales.
Year: 2010