Day 4

Detailed paper information

Back to list

Paper title Multi-sensor ice surface temperature for the Greenland Ice Sheet
Authors
  1. Ioanna Karagali Danish Meteorological Institute Speaker
  2. Magnus Barfod Suhr Danish Meteorological Institute (DMI)
  3. Ruth Mottram Danish Metrological Office
  4. Pia Nielsen-Englyst Technical University of Denmark, DTU SPACE / Danish Meteorological Institute
  5. Gorm Dybkjær Danish Meteorological Institute
  6. Jacob L. Høyer Danish Meteorological Institute (DMI)
Form of presentation Poster
Topics
  • A5. Climate
    • A5.02 The role of Earth Observation in climate services
Abstract text The Greenland Ice Sheet has had a negative mass balance over at least the last two decades, during which there has been a well documented increase in the retreat of the ice sheet. Increased dynamic thinning and lower surface mass balance are roughly equally important mechanisms behind the continuous reduction of the Greenland Ice Sheet, with the latter largely being driven by enhanced melt and run-off rates. In the continuous effort to better simulate the evolution of the Greenland Ice Sheet under different climate change scenarios, models calculate the surface energy budget and convert this to ice surface temperature (IST) in order to calculate melt and run-off. Accurately characterising the ice surface temperature is essential, as it regulates surface melt and run-off through various mechanisms.

Surface temperature monitoring over the polar regions is impeded by harsh environmental conditions, making in situ monitoring challenging and scarce. Space-borne retrievals of ice surface temperature are challenging due to complications from persistent cloud cover, large daily temperature variations and the lack of high quality in-situ observations for validation. Nonetheless, a continuous effort calibrating and harmonising the extended archive of surface temperatures from various sensors has now resulted in comprehensive IST datasets spanning over nearly four decades. A significant part of these datasets are available in satellite processing levels L2 (swath) and L3 (gridded on regular grid) yet with gaps due to cloud cover. Optimally interpolated products offer gap-free fields, typically on a daily basis and while there is a suite of global coverage datasets, few have specifically been developed for the Arctic region.

This study reports from a user case study (UCS) conducted within the ESA CCI LST project. The aim of the UCS was to use the L2 ESA CCI LST products along with the L2 Arctic and Antarctic Ice Surface Temperatures from thermal Infrared satellite sensors (AASTI) v2 dataset, to develop a L4 optimally interpolated, multi-sensor, gap-free, surface temperature field for the Greenland Ice Sheet. The L4 product was produced daily for the year 2012 with a spatial resolution of 0.01 degree latitude and 0.02 degree longitude. Prior to the generation of the gap-free daily fields, the upstream input data were inter-compared and a cold bias for LST CCI MODIS retrievals was identified and corrected against the AASTI dataset. All L2 input data along with the derived product were validated using observations from the PROMICE automated weather stations (AWS) on the Greenland Ice Sheet as well as the IceBridge flight campaigns. L2 AASTI and the L4 OI field shared similar bias and standard deviation values, while MODIS demonstrated a cold bias. The L4 OI fields were used to examine the monthly and seasonal variability of IST during 2012 when a significant melt event occurred. Mean surface temperature for July was around zero for the largest part of the Greenland Ice Sheet, based on the aggregation of 200 to 700 observations depending on the region. Melt days, defined as days when IST was -1°C or higher, ranged between 5 and 10 for the central part of the Greenland Ice Sheet and exceeded 30 for the middle and lower zones in the periphery of the ice sheet. The L4 OI product was assimilated into a surface mass balance (SMB) model of the Greenland Ice Sheet to examine the impact of the multi-sensor, gap free dataset on modelled snowpack properties that account for important effects including refreezing and retention of liquid water for the test year of 2012.