Author(s): Paolo Colosio; Marco Tedesco; Roberto Ranzi

Linked Author(s): Paolo Colosio, Roberto Ranzi

Keywords: Ice sheet melting; Climate change impacts; Sea level rise; Remote sensing; Polar regions

Abstract: The impacts of climate change and global warming are not spatially homogeneous all over the Earth. The largest increase in average temperature has been recorded in the Arctic, with major impact on the cryosphere. The Greenland ice sheet (GrIS) is the largest ice mass in the Northern Hemisphere with a glaciated surface area of about 1,800,000 km² and a thickness up to 3 km and stores a freshwater volume of about 2,900,000 km³, enough to contribute to sea level rise (SLR) by about 7.2 m. According to gravitational data measured by the Gravity Recovery and Climate Experiment (GRACE) satellite mission, between 2002 and 2016, the GrIS lost mass at an average rate of 278±11 Gtons/y, increasing the mean sea level of ~7.9 mm per decade, accelerating of about 21.9 Gtons/y. At these melting and SLR rates, it is urgent to implement adaptation measures (e.g. surface water and groundwater management) in the most vulnerable areas such as Venice, the Netherlands, Bangladesh and Vietnam to face the impacts on coastal processes as floods and saltwater intrusion. The GrIS total mass loss can be split in half between surface processes and dynamic processes. Surface melting affects the surface mass balance by exporting surface meltwater to the ocean and modules dynamics processes, supraglacially, englacially and subglacially. Passive microwave (PMW) brightness temperatures (Tb) are an irreplaceable tool in monitoring seasonal and interannual changes of surface melting over large areas such as the GrIS because of the capability to collect data in all-weather conditions. The major limitation was the slightly coarse spatial resolution (25 km). Here, we use a passive microwave dataset released through the NASA MeASUREs program at the enhanced resolution of 3.125 km to map surface melting over the GrIS. By means of measured (from automatic weather stations) and modelled (MAR regional climate model) data, we found that a dynamic threshold based on the outputs of the electromagnetic model MEMLS is the most suitable, among the assessed algorithms, in detecting sporadic and persistent melting. We found that, during the reference period 1979 – 2019 (1988 – 2019), over the GrIS surface melting in terms of extension, up to 6.9% (3.6%) of the total ice sheet surface extent per decade, and duration, up to 4.5 (2.9) days per decade. Moreover, the melting season has begun ~4 (2.5) days earlier and ended ~7 (3.9) days later every decade. Through a semi-variogram approach, we found that the new product better capture the spatial autocorrelation of surface melting, offering the opportunity to validate modelled melt extent at high resolution and potentially to assimilate this data in climate models to have better projections.

DOI: https://doi.org/10.3850/IAHR-39WC252171192022366

Year: 2022