85th AMS Annual Meeting

Wednesday, 12 January 2005: 2:00 PM
Assimilation of Multi-Satellite High Resolution Sea Surface Temperatures for a Real-Time Local Analysis and Forecasting System
Corey G. Calvert, Florida Institute of Technology, Melbourne, FL; and S. M. Lazarus, P. Santos, D. W. Sharp, P. Blottman, and S. Spratt
Poster PDF (2.6 MB)
The Florida sea breeze and associated deep convection has been studied extensively over the past several decades. However, despite the importance of the sea breeze to Florida weather and climate, operational models use a relatively coarse (50km) SST analysis provided by NCEP's Real-Time Global Sea Surface Temperature (RTG-SST) analysis. Relatively significant SST gradients can be found in the coastal waters to the east of the Florida peninsula a result of a combination of factors including: the Gulf Stream's proximity, intermittent upwelling, and the presence of a relatively cool tongue of water that separates the warmer Gulf Stream from the Florida peninsula north of West Palm Beach. The NWS in Melbourne, FL is cycling (4 times/day), at a horizontal resolution of 4 km, the Advanced Regional Prediction System (ARPS) and its analysis component, the ARPS Data Analysis System (ADAS). We have begun preliminary work on creating a high-resolution SST analysis to replace the RTG-SST analysis currently used to initialize the ARPS SSTs. Various statistics (for the Florida ARPS domain) from the GOES-12 and Moderate Resolution Imaging Spectroradiometer (MODIS) SST data are being examined to determine the quality, variability and reliability of both data sets. Although the MODIS temporal frequency is limited, its high spatial resolution (1 km) and the near real-time availability of data from both the AQUA and TERRA platforms (for which there are a maximum of 4 MODIS passes daily) makes it a viable candidate for operational SST analyses. Monthly, comparisons between the GOES-12 and NDBC buoys in the Florida ARPS domain are quite good - within 0.5C during the winter and 0.3C during the summer. Monthly GOES-12 SST variability peaks during the winter in the near-coastal waters. Despite a data outage at 5-6 UTC (due to a high processing load at NOAA/NESDIS), the GOES-12 SSTs are reliable and accurate in the absence of clouds and solar contamination. Hourly high-resolution (6km) GOES-12 composites and data latency images are currently being delivered to the NWS southern region server in an AWIPS-friendly format.

Supplementary URL: http://my.fit.edu/wx_fit/home/