Uncertainties in Sea Ice Thickness from Radar Altimetry: ESA-CCI Sea Ice ECV Project Round Robin Exercise Results
Djepa, Vera1; Kern, Stefan2; Khvorostovsky, K3; Makynen, Marko4; Rinne, Eero5; Skourup, H6; Sandven, Stein3; Wadhams, Peter1
1Cambridge University, UNITED KINGDOM; 2University of Hamburg,, GERMANY; 3Polar Environment Remote Sensing, NORWAY; 4Finnish Meteorological Institute, Helsinki, FINLAND; 5Finnish Meteorological Institute, Marine research, FINLAND; 6DTU Space National Space Institute, DENMARK
Accurate knowledge of sea ice thickness (SIT) and sea ice volume spatial-temporal distribution is an asset for the correct description and understanding of ocean-ice-atmosphere processes in the Polar region. Both thickness and volume are important parameters to be validated in climate models. Because of the projected change of the sea ice cover it is essential to know SIT as accurately as possible to distinguish sea ice trends from natural variations.
One aim of the ESA CCI sea ice essential climate variable (ECV) project is the generation of a long-term quality controlled SIT data set which shall include physically based information about SIT uncertainty. This data set will be based on satellite radar altimetry which allows measuring the elevation of the ice-snow interface above the water surface: the sea ice free-board, SIF. The uncertainty of parameters required to convert SIF into SIT: snow and ice density, snow depth, add to the measurement uncertainty of SIF. To find the optimal procedure and data combination to retrieve SIT from SIF, and to quantify SIT uncertainty contributions by the input parameters is the goal of the ESA CCI sea ice ECV project Round Robin Exercise which results will be presented here.
Backbone of this exercise is the Round Robin Data Package, RRDP. The RRDP includes co-located data at 100 km grid resolution of sea ice draft (SID), measured by moored and submarine Upward Looking Sonar; SIF, SIT and SID, measured by and obtained from ERS1/2 and ENVISAT radar altimeters; Warren climatology of snow depth (SD) and density; snow free-board, SD, and SIT from Operation Ice Bridge data; snow free-board and SIF from ASIRAS and ALS observations; and SD from Advanced Multichannel Scanning Radiometer. Where given SIT is calculated from SIF and SD or snow free-board and SD assuming isostacy and constant ice and water densities.
The RRDP will be used to quantify the uncertainty of the 100 km grid resolution SIT product based on the above mentioned input parameters and uncertainties; the latter will be carefully reviewed and discussed. We will present results of inter-comparison of the different snow depth data sets in the RRDP to assess their quality and their impact on retrieved SIT. We will inter-compare SID derived from altimetry and SID measured by Upward Looking Sonar. We will present results from a sensitivity study using different sea ice densities to derive SIT from SIF. The presentation will conclude with the recommended set of input parameters and the SIF to SIT conversion method which, based on the Round Robin exercise, is most promising to retrieve the targeted long-term SIT time series from satellite radar altimetry.