Sea Ice Concentration Retrieval Using Composite ScanSAR Features in a SAR Data Assimilation Process

IEEE Trans. Geosci. Remote Sensing

Rinne

Sallila

et al. 2020

In this article, ALOS-2/PALSAR-2 dual-polarized [horizontal-transmit-horizontal-receive and horizontal-transmitvertical-receive (HH/HV)] ScanSAR mode L-band synthetic aperture radar (SAR) imagery over an Arctic study area was evaluated for their suitability for operational sea ice (SI) monitoring. The L-band SAR data are studied for the estimation of different SI parameters: SI concentration, SI thickness, SI type, and SI drift. Also, some comparisons with nearly coincident C-band data over the same study area have been made. The results indicate that the L-band SAR data from ALOS-2/PALSAR-2 are very useful for estimating the studied SI parameters and equally good or better than using the conventional operational dual-polarized C-band SAR satellite data.

Section: Discussionmentioning

confidence: 99%

On Suitability of ALOS-2/PALSAR-2 Dual-Polarized SAR Data for Arctic Sea Ice Parameter Estimation

IEEE Trans. Geosci. Remote Sensing

Rinne

Sallila

et al. 2020

“…One approach toward data fusion from radiometer and SAR data for IC estimation has recently been presented in [17]. This approach is based on a process similar to data assimilation.…”

Section: Discussionmentioning

confidence: 99%

A Comparison Between High-Resolution EO-Based and Ice Analyst-Assigned Sea Ice Concentrations

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

Vainio

Marnela

et al. 2015

In the 4th Ice Analyst Workshop 2014, ice analysts made ice concentration (IC) analysis based on two synthetic aperture radar (SAR) images over the Gulf of Bothnia located in the northern Baltic sea. The ice analysts were divided into five work groups and each group assigned ICs to predefined polygons. This exercise gave us a good opportunity to compare the analysis performed by the work groups with the results produced by our dualpol-SAR IC estimation algorithm. The ice analyzes were performed and the algorithm was run in a resolution of 100 m. For reference, we also included the polygon median values produced by the ARTIST sea ice (ASI) radiometer-based algorithm producing the ICs in a 3.125-km grid. This kind of comparisons are very important to evaluate the automated algorithms with respect to the human interpretation. Human interpretation available in the ice charts is currently considered as the best available IC information. Index Terms-C-band, ice analysis, ice chart, ice concentration (IC), synthetic aperture radar (SAR). Juha Karvonen (M'xx) received the M.Sc. (Tech.), Lic. Tech., and Dr. Sc. (Tech.) degrees in information and computer sciences and digital signal processing from Helsinki university of Technology (Aalto University since

“…A method for ice classification into four ice classes and open water by combining Special Sensor Microwave/Imager (SSM/I) radiometer ice concentration and SAR data was introduced in [29]. Another approach toward data fusion from radiometer and SAR data for SIC estimation has recently been presented in [30] and [31]. This approach is based on a process similar to data assimilation.…”

Section: Introductionmentioning

confidence: 99%

Baltic Sea Ice Concentration Estimation From C-Band Dual-Polarized SAR Imagery by Image Segmentation and Convolutional Neural Networks

IEEE Trans. Geosci. Remote Sensing

2022

In this study application of convolutional neural networks (CNNs) preceded by synthetic aperture radar (SAR), image segmentation for sea ice concentration (SIC) estimation over the Baltic Sea from dual-polarized C-band SAR imagery is studied. Three algorithm variants were studied and trained using FMI ice chart SIC or a synthetic SIC dataset with different SIC values generated by combining pure open water and sea ice blocks by applying binary masks. The first two algorithm variants were trained using only open water and fully ice-covered patches, based on the FMI ice charts, and they had a similar CNN structure. These two algorithm variants differ only in deriving the segmentwise SIC from the CNN output. In the third algorithm, variant synthetic SIC data derived as a mixture of open water and fully ice-covered ice patches according to the ice charts were used in training. The estimation results were evaluated with respect to the FMI ice chart SIC for an independent test dataset. The results were very encouraging for operational purposes and significantly better than for our earlier SAR-based SIC estimation algorithms. The algorithm version trained with synthetic SIC data clearly outperformed the two other algorithm versions and our earlier SIC estimation results based on dual-polarized SAR, using independent FMI ice chart SIC as a reference.