This page lists past seminars and presentations by STAR
scientists and visiting scientists. These seminars include the STAR
Science Forum and similar events. Presentation materials for
seminars will be provided when available.
Title: Using a Generalized Additive Model to Compute Bias-corrected Near-surface Bulk Salinities from Satellite-derived Skin Salinities in the Arctic Ocean and Subarctic Seas
Abstract: This study addresses limitations in Arctic Ocean salinity measurements by utilizing in situ and satellite data, employing a machine-learning approach (Generalized Additive Model; GAM), to convert satellite-derived skin salinity to near-surface (0-5 m) bulk salinity. This research addresses satellite salinity high-latitude retrieval biases, enables the assimilation of those high-latitude satellite salinity observations into numerical modeling, and contributes to validating, verifying, and operationalizing the National Oceanographic and Atmospheric Administration's Unified Forecast System's global coupled model.
Abstract: Satellite constellations such as Sentinel-2A and -2B, Sentinel-3A and -3B, and Planet's PlanetScope constellation offer increased temporal resolution while maintaining spatial, spectral, and radiometric resolutions. For most satellite constellations currently in orbit, platforms are launched either in a group or individually, typically across several years. This increases sampling frequency throughout the satellite mission's lifespan and presents the opportunity to observe more extreme events. When assessing long-term trends or year-over-year change, increased sampling frequency can lead to observed changes that are incorrectly attributed to changes in environmental conditions. This study uses water quality data from the Copernicus Sentinel-3 satellite series to assess temporal aggregation methods for multi-platform satellite missions and their impact on resulting data distributions and change assessments. Temporal aggregation via the maximum data value and via the median data value were compared via the Wilcoxon signed-rank test for a simulation study and for water quality data produced by the Cyanobacteria Assessment Network (CyAN). Next, trends in water quality data were assessed for each temporal aggregation approach using the seasonal Mann-Kendall test for trend and associated Thiel-Sen slope. Results can inform large-scale, long-term water quality monitoring efforts and applications that combine multiple satellite missions, including sensor agnostic workflows.
Remote Access: meet.google.com/uco-uboz-cmk (US) +1 406-838-3189 PIN: 768 242 663#Slides, Recordings Other Materials: available 24-48 hours following the seminar at this link: https://www.star.nesdis.noaa.gov/star/PastSeminars.php
