Arctic Sea Ice Thickness Estimation from CryoSat-2 Satellite Data Using Machine Learning-Based Lead Detection

Abstract

Satellite altimeters have been used to monitor Arctic sea ice thickness since the early 2000s. In order to estimate sea ice thickness from satellite altimeter data, leads (i.e., cracks between ice floes) should first be identified for the calculation of sea ice freeboard. In this study, we proposed novel approaches for lead detection using two machine learning algorithms: decision trees and random forest. CryoSat-2 satellite data collected in March and April of 2011-2014 over the Arctic region were used to extract waveform parameters that show the characteristics of leads, ice floes and ocean, including stack standard deviation, stack skewness, stack kurtosis, pulse peakiness and backscatter sigma-0. The parameters were used to identify leads in the machine learning models. Results show that the proposed approaches, with overall accuracy >90%, produced much better performance than existing lead detection methods based on simple thresholding approaches. Sea ice thickness estimated based on the machine learning-detected leads was compared to the averaged Airborne Electromagnetic (AEM)-bird data collected over two days during the CryoSat Validation experiment (CryoVex) field campaign in April 2011. This comparison showed that the proposed machine learning methods had better performance (up to r = 0.83 and Root Mean Square Error (RMSE) = 0.29 m) compared to thickness estimation based on existing lead detection methods (RMSE = 0.86-0.93 m). Sea ice thickness based on the machine learning approaches showed a consistent decline from 2011-2013 and rebounded in 2014.

Bohai Sea Ice Parameter Estimation Based on Thermodynamic Ice Model and Earth Observation Data

“…Radar altimetry [34][35][36] is not suitable for a high-resolution SIT estimation of relatively thin sea ice. The SIT estimates based on radar altimetry, such as CryoSat-2, include many uncertainties, especially in the freeboard-to-thickness conversion [84], which typically contributes more to the SIT estimate than does the typical ice thickness in the Bohai Sea.…”

“…Space-borne remote sensing of SIT can be conducted based on Archimedes' law and satellite radar or laser altimeter measurements of sea ice freeboard [34][35][36]. However, this method produces large relative errors for thin-ice [34], and is thus not suitable for the Bohai Sea where only thin first year ice occurs.…”

Parameters Comparsion of Leads Detection in Arctic Sea Ice Using Cryosat-2 Waveform Data

“…Then, on the basis of the above conditions, the parameters SKU, SSK are added successively. According to Lee (2016), the constrain conditions are SKU > 80, SSK > 9, respectively. Finally, on the basis of the combination of PP&SSD parameters, SKU and SSK are added simultaneously.…”

“…After comparing the results with MODIS images, it turns out it's best to distinguish between the leads and sea ice with the maximum parameter. ESA (2014) give the equations of parameters and Lee (2016) use the equations of SSD, SKU, SSK, PP and the value range of sea ice, leads and sea water corresponding to different parameters. Two other methods are proposed, See 5.0 decision trees and random forest, the accuracy of the two is pretty high.…”

“…Simple thresholding methods might not perfectly distinguish leads from ice floes because parameters, such as PP, SSD, SSK, SKU and backscatter sigma-0 typically contain aliasing between leads and ice floes, which can result in large errors and uncertainties in sea ice thickness estimates. In this paper, four parameters from L1B waveform data of CryosSat-2 satellite are extracted, according to Laxon (2013) and Lee (2016), five combinations (PP, PP&SSD, PP&SSD&SKU, PP&SSD&SSK, PP&SSD&SKU&SSK) are proposed to detect the leads, and the results are combined with MODIS images.…”