The research focuses on ice deformation, the process controlling how glaciers and ice sheets shift, which has been affected by climate change.
University of Otago department of geology lead author Dr Sheng Fan said estimating sea level rise depended on a model called flow law, a mathematical equation describing how ice flows.
"We need a more precise flow law so we can reduce prediction errors, especially with the way climate change is progressing."
Using advanced statistical methods, researchers pooled 70 years of global data to improve on the commonly used flow laws.
Researchers found past models underestimated the complexity of ice deformation, resulting in less reliable forecasts.
By incorporating factors such as ice grain size, the new approach provided a clearer view of future melting.
University of Otago department of geology Prof David Prior said many factors contributed to sea level rise and the future of the ice sheet was probably the biggest uncertainty.
"This study illustrates that we need to describe the behaviour of the ice much more precisely, particularly if we want to use ice sheet modelling as a predicting tool.
"If we want our predictions of ice movement over the next few decades to be robust, we need to get the physics right."
Both field observations and laboratory experiments were needed to understand how ice behaved over time.
"Measurements of ice velocity from satellites, for example, can be used to tell us about the physics of ice flow and have the advantage that the measurements are of real ice under natural conditions," Prof Prior said.
However, working out the physics required making assumptions — for example, about how ice slides at the bottom and sides, estimating key data such as the gravity slope driving motion and averaging properties across the full thickness of the ice sheet.
The research, published in Nature Geoscience, was supported by the Royal Society Te Apārangi Marsden Fund and Antarctica New Zealand.
It featured researchers from the universities of Cambridge, Pennsylvania and Maryland as well as the University of Otago.
