CaltechAUTHORS
  A Caltech Library Service

Deep Learning for Predicting Significant Wave Height From Synthetic Aperture Radar

Quach, Brandon and Glaser, Yannik and Stopa, Justin Edward and Mouche, Alexis Aurélien and Sadowski, Peter (2020) Deep Learning for Predicting Significant Wave Height From Synthetic Aperture Radar. IEEE Transactions on Geoscience and Remote Sensing . ISSN 0196-2892. (In Press) https://resolver.caltech.edu/CaltechAUTHORS:20200724-100654154

[img] PDF - Accepted Version
See Usage Policy.

23Mb

Use this Persistent URL to link to this item: https://resolver.caltech.edu/CaltechAUTHORS:20200724-100654154

Abstract

The Sentinel-1 satellites equipped with synthetic aperture radars (SARs) provide near-global coverage of the world's oceans every six days. We curate a data set of collocations between SAR and altimeter satellites and investigate the use of deep learning to predict significant wave height from SAR. While previous models for predicting geophysical quantities from SAR rely heavily on feature-engineering, our approach learns directly from low-level image cross-spectra. Training on collocations from 2015 to 2017, we demonstrate on test data from 2018 that deep learning reduces the state-of-the-art root mean squared error by 50%, from 0.6 to 0.3 m when compared to altimeter data. Furthermore, we isolate the contributions of different features to the model performance.


Item Type:Article
Related URLs:
URLURL TypeDescription
https://doi.org/10.1109/tgrs.2020.3003839DOIArticle
ORCID:
AuthorORCID
Quach, Brandon0000-0001-9848-7483
Glaser, Yannik0000-0001-7217-9749
Stopa, Justin Edward0000-0002-7477-8224
Mouche, Alexis Aurélien0000-0003-1250-4436
Additional Information:© 2020 IEEE. Manuscript received February 14, 2020; revised May 22, 2020; accepted June 8, 2020. This work was made possible, thanks to SAR data access granted by ESA projects: Sentinel-1 A Mission Performance Center (4000107360/12/I-LG) and Sentinel-1 Ocean Study (S1-4SCI-16-0002). All Sentinel-1 L2 data used in this study can be obtained from the Copernicus Data Hub (cophub.copernicus.eu). The buoy data can be obtained from the respective centers: NDBC (nodc.noaa.gov/BUOY/), MEDS (medssdmm.dfo-mpo.gc.ca), and OceanSITES (www.oceansites.org/). NSF Ocean Observatories Initiative Data Portal, http://ooinet.oceanobservatories.org, Surface Wave Spectra (CE02SHSM, CE04OSSM, CE07SHSM, CE09OSSM, GA01SUMO, GI01SUMO, GS01SUMO, CP01CNSM: -SBD1205-WAVSSA000) data from September 10, 2014 to July 31, 2018. Downloaded on July 14, 2018. The altimetry data was sourced from the Integrated Marine Observing System (IMOS) - IMOS is a national collaborative research infrastructure, supported by the Australian Government. IMOS 2014–2018, IMOS - SRS Surface Waves Sub-Facility - altimeter wave/wind, https://portal.aodn.org.au, accessed January 24, 2018. The authors would like to thank NVIDIA for a hardware grant to PS. The technical support and advanced computing resources from the University of Hawai‘i Information Technology Services Cyberinfrastructure are gratefully acknowledged.
Funders:
Funding AgencyGrant Number
NVIDIAUNSPECIFIED
Subject Keywords:CWAVE, deep learning, machine learning, neural networks, Sentinel-1, significant wave height, synthetic aperture radar (SAR)
Record Number:CaltechAUTHORS:20200724-100654154
Persistent URL:https://resolver.caltech.edu/CaltechAUTHORS:20200724-100654154
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:104562
Collection:CaltechAUTHORS
Deposited By: George Porter
Deposited On:24 Jul 2020 20:02
Last Modified:24 Jul 2020 20:02

Repository Staff Only: item control page