Introduction
The Antarctic Ice Sheet is one of the world’s greatest freshwater storages. A warming climate has led to accelerated mass loss, mainly from its Western part  (WAIS), and its contribution to sea level rise has has steadily increased over the last decade. The aim of RATES is to quantify and ascertain rigorous uncertainty estimates on this contribution. 

Mass balance
Mass and height changes on the ice sheets are determined mainly by three processes: ice dynamics, surface mass balance, and land movement.  Surface mass balance (SMB) subsumes snowfall (or precipitation in general), melt, and runoff. Land movement can be caused through either elastic or viscoelastic processes: When large ice masses melt away, the solid earth springs back into place where is was pushed down by ice masses before. This can either happen instantaneously - if only the elastic upper layer, the crust, is concerned - or over thousands of years as a reaction of the fluid upper mantle flowing back into place. In the latter case, this process is called glacial isostatic adjustment - GIA. Another process that plays a role in determining mass balance is firn compaction, the densification of past years’ snow (firn). So overall, we are dealing with five different processes: ice, SMB, GIA, elastic rebound of the crust, and firn compaction. It is worth pointing out that the two latter processes result in height changes, but do not affect mass. Yet all five need to be known or estimated if we are to obtain a reliable estimate for Antarctica's contribution to sea level rise.


Combining all data
Our data can identify total height and mass changes, but can not necessarily discriminate between the different causes. The aim of the RATES project is to produce error-bounded trends of Antarctic ice mass balance through a rigorous combination of different data sets - altimetry, satellite gravimetry, and GPS. Our agenda is to use as few forward models as possible for our estimates. We are employing a hierarchical model (HM) statistical framework which simultaneously estimates the "most likely" combination of ice and SMB changes and viscoelastic rebound (GIA) at any given point - together with a probability distribution and a standard error. See the section 'Project details' for more details.

The team


Professor Jonathan Bamber    Dr Nana Schoen  Dr Andrew Zammit Mangion