***************************************************************** 2008-08-08 11:28:00 +0000: Journal for accession number: 0044419 Created on: 2008-08-08 11:28:00 +0000 By: Kenneth Casey, OC1, SSMC3 4853, +13017133272 x133 Source: This accession contains the Coral Reef Temperature Anomaly Database (CoRTAD), a Global, 4 km, Sea Surface Temperature and Related Thermal Stress Metrics dataset for 1985-2005 (NODC Accession Number 0044419). It was created at NODC using AVHRR Pathfinder V5 SSTs, in association with Elizabeth Selig (University of North Carolina - Chapel Hill, now Conservation International) and John Bruno (UNC - Chapel Hill). These data were created with support from the NOAA Coral Reef Conservation Program and are freely and openly available. Please acknowledge the use of these data with: "The Coral Reef Temperature Anomaly Database (CoRTAD) was developed by the NOAA National Oceanographic Data Center and the University of North Carolina - Chapel Hill (Selig, E.R., K.S. Casey, and J. Bruno (2008). Global patterns of variability in coral reef temperature anomalies: the importance of fine scale spatial and temporal heterogeneity. Global Change Biology, submitted.) It was provided by the NOAA National Oceanographic Data Center at URL:http://www.nodc.noaa.gov/SatelliteData/Cortad/" The dataset consists of 128 files in HDF4 Scientific Data Set format, which is a self-documenting format. Each of these files represents one "tile", and each tile covers a given region of the Earth. See the image file called TileMap.jpg to see which tiles cover which regions. The globe is divided into 8 rows and 16 columns. The file names follow this convention: cortad_rowYY_colXX.hdf where row YY varies from 00 to 07 and column XX varies from 00 to 15. Each tile file contains the following variables: WeeklySST: Gappy weekly SST time series for 1985-2005 Latiude: A vector of latitudes, South is negative Longitude: A vector of longitudes, West is negative Time: A vector of times for each time step in the 3-dimensional data matrices, in units of Center Year and Day of Year (YYYYDDD) MedfillSST: Partially gap-filled weekly SST time series after median fill for 1985-2005 FilledSST: Gap-free weekly SST time series for 1985-2005 after median spatial and temporal pchip interpolation ClimSST: Climatological SST based on weekly SSTs for 1985-2005, created using harmonics approach. Harmonics: The five coefficients determined using the fit harmonics approach, from which ClimSST is evaluated. SSTA: Weekly SST Anomalies SST_Stats: Minimum, Maximum, Standard Deviation, and Mean of SST time series SSTA_Stats: Minimum, Maximum, Standard Deviation, Mean, and Mean of absolute values of SST Anomaly (SSTA) time series SSTA_Frequency: Frequency of SST Anomalies >=1 deg C over previous 52 weeks SSTA_Frequency_Stats: Maximum, Standard Deviation, and Mean of SST Anomaly Frequency (SSTA_Frequency) time series SSTA_DHW: Sum of SST Anomalies >=1 deg C over previous 12 weeks (SSTA Degree Heating Week) SSTA_DHW_Stats: Maximum, Standard Deviation, and Mean of SST Anomaly Degree Heating Week (SSTA_DHW) time series TSA: Thermal Stress Anomalies (defined as SST minus Maximum Weekly Mean Climatological SST TSA_Stats: Maximum, Standard Deviation, and Mean of Thermal Stress Anomaly (TSA) time series TSA_Frequency: Frequency of Thermal Stress Anomalies >=1 deg C over previous 52 weeks TSA_Frequency_Stats: Maximum, Standard Deviation, and Mean of Thermal Stress Anomaly Frequency (TSA_Frequency) time series TSA_DHW: Sum of Thermal Stress Anomalies >=1 deg C over previous 12 weeks (TSA Degree Heating Week) TSA_DHW_Stats: Maximum, Standard Deviation, and Mean of Thermal Stress Anomaly Degree Heating Week (TSA_DHW) time series AllBad: A matrix showing which pixels are always missing. These missing pixels are not gap-filled and no analyses take place at those locations Land: A matrix showing which pixels are on land. No analyses take place at those locations. NumberGood: A matrix showing how many of the original SST observations are valid prior to any gap-filling. Note scales and offsets must be applied in some cases as described in the HDF files and in the table below. To apply the SCALE and OFFSET, use a simple linear relationship: Geophysical Value = Pixel_Value * SCALE + OFFSET SCALE OFFSET WeeklySST: 0.075 -3.000 Latitude: 1.000 0.000 Longitude: 1.000 0.000 Time: 1.000 0.000 MedfillSST: 0.075 -3.000 FilledSST: 0.075 -3.000 ClimSST: 0.075 -3.000 Harmonics: 0.075 0.000 * * NOTE: To convert the amplitudes (layers 1 and 3) to deg C, use a scale of 0.075 and an offset of 0, but to convert the long term mean (layer 5) to deg C use a scale of 0.075 and an offset of -3.0 SSTA: 0.001 -7.500 SST_Stats: 1.000 0.000 SSTA_Stats: 1.000 0.000 SSTA_Frequency: 1.000 0.000 SSTA_Frequency_Stats: 1.000 0.000 SSTA_DHW: 0.010 0.000 SSTA_DHW_Stats: 1.000 0.000 TSA: 0.075 0.000 TSA_Stats: 1.000 0.000 TSA_Frequency: 1.000 0.000 TSA_Frequency_Stats: 1.000 0.000 TSA_DHW: 0.075 0.000 TSA_DHW_Stats: 1.000 0.000 AllBad: 1.000 0.000 Land: 1.000 0.000 NumberGood: 1.000 0.000 The other JPEG files in this directory are for illustration purposes and contain: currmax.jpg - Map of Maximum SST at each location in the CoRTAD currmean.jpg - Map of Mean SST at each location in the CoRTAD currmin.jpg - Map of Minimum SST at each location in the CoRTAD A higher resolution TIFF version of each of these is also provided. An FGDC metadata record in XML format is contained in the file CoRTAD_FGDC.xml.