Ship-Image map derived from a Terra MODIS image from October 20, 2003 (01:35 UT acquisition time), showing the course of the ARISE cruise (red line) between September 25 and October 21, 2003, and tracks of helicopter flights with validation data. ‘Heli Flt 2’ (green), was acquired on October 3; ‘Heli Flt 3’ (blue), October 3 ‘Heli Flt 4’ (yellow), October 8 ‘Heli Flt 5’ (cyan), October 8 ‘Heli Flt 6’ (magenta), October 19 ‘Heli Flt 7’ (maroon), October 20 ‘Heli Flt 8’ (dark green), October 20 Image shows approximate sea ice conditions for only the last three helicopter flights. 44A053 Validation of AVHRR and MODIS Ice Surface Temperature Products Using In Situ Radiometers Terry Haran & Ted Scambos National Snow and Ice Data Center, CIRES, University of Colorado, Boulder, CO, USA Rob Massom Antarctic Climate and Ecosystems CRC, University of Tasmania, Hobart TAS 7001 Australia ftp://sidads.colorado.edu/pub/ppp/modis-avhrr/SeaIcePoster2005_top.ppt and SeaIcePoster2005_bot.ppt Abstract Ship-borne and airborne infrared radiometric measurements during the Antarctic Remote Ice Sensing Experiment (ARISE) cruise of September-October, 2003 permit in situ validation studies of two satellite-based ice surface skin temperature algorithms: the AVHRR Polar Pathfinder Ice Surface Temperature; and the MODIS Sea Ice Surface Temperature. Observations of sea ice from the Aurora Australis ship's rail using a KT radiometer were conducted between September 25 and October 21 during clear-sky overflights by AVHRR (41 passes) and MODIS (17 passes) on their respective satellite platforms. Data from both sensors show highly linear fits to one minute integrated radiometer spot measurements spanning the range 245 K to 270 K with a ±2.8°C (AVHRR) and ±2.0°C (MODIS) variation. There was no significant offset. Helicopter observations made with a KT radiometer on three dates (October 8, 19th, and 20th) provided more data, but over a more limited sea ice skin temperature range (252 K to 268 K). Comparison of MODIS and AVHRR algorithms directly, with both images acquired during a helicopter flight, indicates very high correlation and near-unity slope for the two satellite-based algorithms. Ship air temperature data during the validation indicated moderate to strong inversions over sea ice under clear skies. Inversions of 2°C to 15°C were observed between the KT-19 and the ship-board air temperature at 21m elevation. These formed and decayed rapidly (tens of minutes) as clouds moved out of and into the zenith area. ARISE Cruise and Helicopter Flights In Situ Radiometers Downward looking KT infrared radiometer mounted inside helicopter. Adjustable look angle KT infrared radiometer mounted on ship’s rail. Rail-mounted KT vs. AVHRR and MODIS Validation of AVHRR Polar Pathfinder (a) ice surface temperature and MODIS sea ice surface temperature (b) using ship- rail-mounted KT data. Ranges of image-grid pixel values at each calibration site shown by triangles above and below the mean value. Error of KT measurement shown by width of box symbol (±0.3°C). Best fit line, with equation, shown for both sensors (thick dashed line); 1:1 line (solid), with lines at 2.5°C above and below the 1:1 (thin dotted lines) shown for reference. Flight 5 Helicopter-mounted KT vs. AVHRR and MODIS Helicopter validation flight 5 of October 8. a: Regional sea ice extent and clouds shown in MODIS Channel 1 image (250m pixel size) acquired during flight 5 (Aqua 08:24 UT). b: Close-up of helicopter ground track from a. c: MODIS-derived sea ice surface temperature and cloud mask image. d: NOAA-16 AVHRR-derived ice surface temperature is shown (acquisition time 08:39 UT). A NOAA-12 image was also acquired during the flight (not shown, acquisition time 08:48). e: Along-track KT skin temperature variations with time are shown, as well as Aqua MODIS, AVHRR NOAA-16, and NOAA-12 values at the track locations. Vertical colored bars show time range of ±400 sec, used below for validation runs. Note drift in temperature over ice floes in the black KT-19 track due to diurnal temperature change during the 5640 second = 94 minute flight. Solid black line shows linear fit to KT temperatures acquired between seconds and (= 2240 seconds or about 37 minutes) which had a slope of °C/min or -1.12°C in 15 minutes or -2.76°C in 37 minutes.