207
The NOAA Daily Optimum Interpolation SST: Status and planned improvements

- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner
Monday, 5 January 2015
Viva F. Banzon, NOAA/NESDIS/NCDC, Asheville, NC; and D. Saunders, A. Burden, C. Liu, and M. Urzen

Handout (2.3 MB)

The NOAA Daily Optimum Interpolation Sea Surface Temperature (DOISST) on a 1/4° spatial grid, produced at the National Climatic Data Center, was developed to support climate modeling and research. It has a finer temporal and spatial resolution than its precursor, the 1° weekly OISST, that is used for seasonal monitoring at NCEP. Near real time daily updates to DOISST are also gaining use as an ancillary dataset for current satellite missions and for ecological monitoring products. OISSTs are produced by blending in situ and satellite SSTs, and include bias correction steps to take into account platform differences. The version of DOISST currently in production uses satellite SSTs from the Advanced Very High Resolution Radiometer (AVHRR) and is thus called AVHRR_ONLY. Another version called AVHRR+AMSR includes microwave SSTs but is available only from 2002 to 2011. At present, the planned improvements to AVHRR_ONLY, in line with the Climate Data Records Program requirements, involve code refactoring, followed by a reprocessing. The refactoring is transparent to users but is intended to increase code accessibility and reduce platform dependence. Once the refactoring is completed, a reprocessing will be done using newer and extended versions of the input datasets (Pathfinder 5.2 SSTs up to 2012, GSFC sea ice concentrations up to 2012; ICOADS 2.5 up to 2007). This is expected to increase DOISST temporal consistency. Operational data will be still used for daily updates, as has been done in the past. The switch to operational data can have noticeable effects, so some changes will address this issue. For example, proxy SSTs at the marginal ice zone will be generated using a higher resolution operational NCEP analysis to minimize a jump in sea ice cover when transitioning from GSFC ice. Differences between the two types of updates (i.e., preliminary or near real time, produced with 1-day latency, and the final DOISST, produced after 15 days) will also be discussed. Production of the version of DOISST that includes microwave data will resume once these tasks described here are completed, but will be renamed since other microwave sensors will be used to extend the time series. Because of the all-weather coverage provided by microwave data, this other version may offer a better ancillary field for use in operational satellite algorithms.

Supplementary URL: http://www.ncdc.noaa.gov/sst