Search for a command to run...
Abstract. Advances in remote sensing of sea ice over the past two decades have resulted in a wide variety of satellite-derived sea ice thickness data products becoming publicly available. Selecting the most appropriate product is challenging given objectives range from incorporating satellite-derived thickness information in operational activities, including sea ice forecasting, routing of maritime traffic, and search and rescue, to climate change analysis, longer-term modeling, prediction, and future planning. Depending on the use case, selecting the most suitable satellite data product can depend on the region of interest, data latency, and whether the data are provided routinely, for example via a climate or maritime service provider. Here we examine a suite of current sea ice thickness data products, collating key details of primary interest to end users. We assess sea ice thickness observations derived from sensors onboard the CryoSat-2 (CS2), Advanced Very-High-Resolution Radiometer (AVHRR) and Soil Moisture and Ocean Salinity (SMOS) satellites. We evaluate the satellite-only observations with independent ice draft and thickness measurements obtained from the Beaufort Gyre Exploration Project (BGEP) upward looking sonars (ULS) and Operation IceBridge, respectively. We find a number of key differences among data products, but find that products utilizing CS2-only measurements are reliable for sea ice between ~ 0.5 m and 4 m. Among those compare, a blended CS2-SMOS product was the most reliable for thin ice. Ice thickness distributions at the end of winter appeared realistic when compared with independent ice draft measurements, with the exception of those derived from AVHRR. We noticed disagreement among the products in mean thickness trends especially in the winters of 2012–2013 and 2016–2017. Regional comparisons reveal large differences in ice thickness between products, particularly in the Kara Sea, an area of considerable ship traffic.
DOI: 10.5194/tc-2018-197