Sea ice remote sensing using spaceborne global navigation satellite system reflectometry

• Main Author: Yan, Qingyun
• Format: Text
• Language: English
• Published: Memorial University of Newfoundland 2019
• Subjects: Sea ice
• Online Access: https://dx.doi.org/10.48336/v6mz-zq52; https://research.library.mun.ca/14413/

In this research, the application of spaceborne Global Navigation Satellite System- Reflectometry (GNSS-R) delay-Doppler maps (DDMs) for sea ice remote sensing is investigated. Firstly, a scheme is presented for detecting sea ice from TechDemoSat-1 (TDS-1) DDMs. Less spreading along delay and Doppler axes is observed in the DDMs of sea ice relative to those of seawater. This enables us to distinguish sea ice from seawater through studying the values of various DDM observables, which describe the extent of DDM spreading. Secondly, three machine learning-based methods, specifically neural networks (NNs), convolutional neural networks (CNNs) and support vector machine (SVM), are de- veloped for detecting sea ice and retrieving sea ice concentration (SIC) from TDS-1 data. For these three methods, the architectures with different outputs (i.e. cate- gory labels and SIC values) are separately devised for sea ice detection (classification problem) and SIC retrieval (regression problem) purposes. In the training phase, dif- ferent designs of input that include the cropped DDM (40-by-20), the full-size DDM (128-by-20), and the feature selection (FS) (1-by-20) are tested. The SIC data ob- tained by Nimbus-7 SMMR and DMSP SSM/I-SSMIS sensors are used as the target data, which are also regarded as ground-truth data in this work. In the experimental stage, CNN output resulted from inputting full-size DDM data shows better accuracy than that of the NN-based method. Besides, performance of both CNNs and NNs is enhanced with the cropped DDMs. It is found that when DDM data are adequately preprocessed CNNs and NNs share similar accuracy. Further comparison is made between NN and SVM with FS. The SVM algorithm demonstrates improved accuracy compared with the NN method. In addition, the designed FS is proven to be effective for both SVM- and NN-based approaches. Lastly, a reflectivity (Γ)-based method for sea ice thickness (SIT) retrieval is proposed. SIT is calculated from TDS-1 Γ data, and verified with two sets of reference SIT data; one is from University of Hamburg and obtained by the Soil Moisture Ocean Salinity (SMOS) satellite, and the other is the combined SMOS/Soil Moisture Active Passive (SMAP) measurements from University of Bremen. This analysis is performed on the data with sea ice thickness less than 1 m. Through comparison, a good consistency between the derived TDS-1 SIT and the reference SIT was obtained, with a correlation coefficient (r) of 0.84 and a root-mean-square difference (RMSD) of 9.39 cm with SMOS, and an r of 0.67 and an RMSD of 9.49 cm with SMOS/SMAP, which demonstrates the applicability of the developed model and the utility of TDS-1 data for SIT estimation. In addition, this method is shown to be useful for improving proposed sea ice detection methods.