Comparison of Near−Surface Air Temperatures from Multiple AWS and MODIS Ice−Surface Temperatures at Summit, Greenland (2008−2014)

Christopher A. Shuman, University of Maryland, Baltimore County at NASA Goddard Space Flight Center, Greenbelt, United States, Dorothy K Hall, NASA Goddard Space Flight Center, Cryospheric Sciences Laboratory, Greenbelt, MD, United States, Nicolo E. DiGirolamo, Science Systems and Applications, Inc., Lanham, MD, United States, Thomas K Mefford, NOAA / ESRL Global Monitoring Division, Boulder, CO, United States and Michael J. Schnaubelt, University of Maryland Baltimore County, Baltimore, MD, United States
Abstract:
We have investigated the stability of the MODerate−resolution Imaging Spectroradiometer (MODIS) ice−surface temperature (IST) product from Terra and Aqua for use as a climate−quality data record. The availability of climate−quality air temperature data (TA) from a NOAA observatory at Greenland’s Summit station as well as more traditional temperature data from nearby automatic weather stations (AWS) has enabled this high−temporal resolution study of MODIS ISTs. During a >6 year period (July 2008 to August 2014), the Terra and Aqua Collection 5 IST values were compared with contemporaneous average TA values from Summit Station’s nominal 2 m temperature sensor data sets (NOAA TAWO, GC-Net and DMI AWS).

This approach has enabled an expected small offset between air and ice sheet surface temperatures (TA > IST) to be investigated over multiple annual cycles. Previous work (Shuman et al., 2014, J. App. Meteo & Clim.) documents a number of factors influencing a progressive ‘cold bias’. This comparison, during 2008−2013, shows that there is a difference of about −0.5°C at the upper end of the NOAA-derived temperature range that increases to as much as −5°C on average at −60°C after some additional cloud filtering of the IST data. The consistency of the comparison results over each year in this study indicates that MODIS provides an alternative platform for remotely deriving surface temperature data, with the resulting IST data most compatible with in situ TA data when temperatures are warm and skies are clear. Finally, the ongoing IST data set should benefit from improved cloud filtering as well as algorithm modifications in Collection 6 to account for the progressive offset from TA at colder temperatures. Additional work will be done when Collection 6 IST data become available.

Back to: Remote Sensing of the Cryosphere III Posters