Monitoring the Arctic Seas: How Satellite Altimetry Can Be Used to Detect Open Water in Sea-Ice Regions

Abstract

Open water areas surrounded by sea ice significantly influence the ocean-ice-atmosphere interaction and contribute to Arctic climate change. Satellite altimetry can detect these ice openings and enables one to estimate sea surface heights and further altimetry data derived products. This study introduces an innovative, unsupervised classification approach for detecting open water areas in the Greenland Sea based on high-frequency data from Envisat and SARAL. Altimetry radar echoes, also called waveforms, are analyzed regarding different surface conditions. Six waveform features are defined to cluster radar echoes into different groups indicating open water and sea ice waveforms. Therefore, the partitional clustering algorithm K-medoids and the memory-based classification method K-nearest neighbor are employed, yielding an internal misclassification error of about 2%. A quantitative comparison with several SAR images reveals a consistency rate of 76.9% for SARAL and 70.7% for Envisat. These numbers strongly depend on the quality of the SAR images and the time lag between the measurements of both techniques. For a few examples, a consistency rate of more than 90% and a true water detection rate of 94% can be demonstrated. The innovative classification procedure can be used to detect water areas with different spatial extents and can be applied to all available pulse-limited altimetry datasets.

Lead Detection in Polar Oceans—A Comparison of Different Classification Methods for Cryosat-2 SAR Data

“…The unsupervised classification method developed by Müller et al [11] is believed to be the first attempt using this method applied to satellite altimetry over sea-ice areas. There are many different approaches to unsupervised learning including clustering (e.g., K-means, K-medoids [27]), latent variable models (e.g., Hidden Markov Models [28]) and others.…”

“…There are many different approaches to unsupervised learning including clustering (e.g., K-means, K-medoids [27]), latent variable models (e.g., Hidden Markov Models [28]) and others. Müller et al [11] employed the clustering K-medoids algorithm for their unsupervised classification on pulse-limited altimetry waveforms (Envisat and SARAL). In this study, the unsupervised classification method is tested for the first time with SAR altimeter data.…”

“…The six waveform features defined by Müller et al [11] are also used for the clustering in this study, with minor changes in some computation parameters:…”

“…This parameter also considers multiple consecutive observations in an attempt to isolate a lead at the nadir position. Müller et al [11] developed an unsupervised classification method based on K-medoids partitional clustering [12] of model waveform features and classification of observation data relative to the model using K-nearest neighbor (KNN) classification [13]. Shen et al [14] aimed to find an optimal classifier-feature assembly for sea-ice classification.…”

“…The aim is to find the best existing CS-2 SAR altimeter lead classification algorithm that provides low false classification. The four classification methods under investigation are (1) the waveform maximum power as introduced, e.g., by Röhrs and Kaleschke [7]; (2) the multi-parameter classification method primarily based on pulse peakiness developed by Ricker et al [16]; (3) a method based on the stack peakiness presented by Passaro et al [10]; and (4) the unsupervised classification method published by Müller et al [11]. This is the first study that compares different classification approaches using exactly the same input data and the same very high resolution (<1 m) ground truth.…”