Assessing concentration uncertainty estimates from passive microwave sea ice products
Abstract
Sea ice concentration is an essential climate variable and passive microwave derived estimates of concentration are one of the longest satellite-derived climate records. However, until recently uncertainty estimates were not provided. Numerous validation studies provided insight into general error characteristics, but the studies have found that concentration error varied greatly depending on sea ice conditions. Thus, an uncertainty estimate from each observation is desired, particularly for initialization, assimilation, and validation of models. Here we investigate three sea ice products that include an uncertainty for each concentration estimate: the NASA Team 2 algorithm product, the EUMETSAT Ocean and Sea Ice Satellite Application Facility (OSI-SAF) product, and the NOAA/NSIDC Climate Data Record (CDR) product. Each product estimates uncertainty with a completely different approach. The NASA Team 2 product derives uncertainty internally from the algorithm method itself. The OSI-SAF uses atmospheric reanalysis fields and a radiative transfer model. The CDR uses spatial variability from two algorithms. Each approach has merits and limitations. Here we evaluate the uncertainty estimates by comparing the passive microwave concentration products with fields derived from the NOAA VIIRS sensor. The results show that the relationship between the product uncertainty estimates and the concentration error (relative to VIIRS) is complex. This may be due to the sea ice conditions, the uncertainty methods, as well as the spatial and temporal variability of the passive microwave and VIIRS products.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2017
- Bibcode:
- 2017AGUFM.C21G1182M
- Keywords:
-
- 0750 Sea ice;
- CRYOSPHERE;
- 1640 Remote sensing;
- GLOBAL CHANGE;
- 4207 Arctic and Antarctic oceanography;
- OCEANOGRAPHY: GENERAL;
- 4262 Ocean observing systems;
- OCEANOGRAPHY: GENERAL