1.1 Objectives of and justification for work. We will define trends and constrain uncertainty in surface temperature of the Greenland Ice Sheet (GIS) from 1981 to the present. Of particular interest are analyses of decade-length trends as well as changes in surface-temperature patterns that may indicate a change in the amount, timing, and spatial extent of surface melting. Trends will be studied in the context of climatological data such as the North Atlantic Oscillation (NAO) index to improve our understanding of the role of warming climate and oscillatory trends on ice-sheet temperatures. We will develop a high-quality, validated climate-data record (CDR), with error bars, of clear-sky surface temperature of the GIS using infrared (IR) measurements and data products from the Advanced Very High Frequency Radiometer (AVHRR) and Moderate-Resolution Imaging Spectroradiometer (MODIS) from 1981 to the present. Issues that contribute to errors or biases in current IR records will be explored and resolved. This CDR will enable surface-flux and time-series studies and will be validated and distributed to the wider scientific and modeling communities.
1.2 Approach. We will use AVHRR and MODIS data to develop a record of daily, clear-sky surface temperature of the GIS, and employ Ice, Cloud, and land Elevation Satellite (ICESat), automatic weather stations (AWS), Landsat Enhanced Thematic Mapper Plus (ETM+) and Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) data to validate and refine the surface-temperature record. Since surface temperatures are generally higher beneath clouds, and IR satellite sensors cannot �see� through clouds, these sensors generally tend to underestimate surface temperature beneath clouds, and during persistently-cloudy periods. To assess the reliability of our IR retrievals, we will study profiles derived from ICESat cloud products overlain on nearly-coincident MODIS surface-temperature pixels. We will compare the result of the MODIS standard cloud mask with the ICESat-derived cloud-optical thickness data to assess �cloud or not cloud,� especially in areas of thin clouds which represent the greatest source of uncertainty. To validate the AVHRR- and MODIS-derived surface temperatures, we will use available AWS and other station data. We will also explore the use of IR surface temperatures derived from ETM+ and ASTER for validation of AVHRR- and MODIS-derived surface temperatures.
1.3 Expected Outcomes of the Project. We will develop an improved understanding of, and document, decade-scale trends in the surface-melt dynamics of the GIS including timing of melt onset, duration and extent of melt, melt-pond evolution, and annual, seasonal and interannual ice sheet surface temperatures to improve our understanding of ice-sheet surface processes. A high-quality, internally-consistent, satellite-derived record of ice sheet surface temperature, with error bars, will be developed using data from AVHRR and MODIS instruments and made available to the community at large. The standard products will be weekly and monthly maps with constrained uncertainties. Observed surface-temperature trends will be explored in the context of broader climatological processes such as the NAO. This CDR can then be utilized in many polar process and mass-balance studies, including modeling studies, that require a spatially-extensive, validated surface-temperature record of the GIS.
