Published on March 5th, 2014 | by Emily Corbett0
Waves in Sea Ice
By Ewan Short, University of Adelaide
This student took part in the 2012/13 AMSI Vacation Research Scholarship program. For more information on this years program please click here
One of the most important uses of mathematics is the modelling of physical systems. This involves taking the physical laws that govern a system, expressing them mathematically, then using a computer simulation to attempt to predict the future state of the system. Modelling is a messy process. Significant assumptions need to be made to simplify the system being modelled, and to allow a computer simulation to be applied. However, as computing power has increased, models have been made increasingly sophisticated, and they now form a key part of modern science.
For example, large scale climate modelling helps substantiate some of the main arguments of organisations like the IPCC for anthropogenic global warming (IPCC 2007). These incredibly complex models are built out of smaller sub models, each responsible for a particular aspect of the climate system. A typical climate model might include ocean, atmosphere and sea ice component models all feeding information to one another to build a picture of the whole climate system.
One process currently neglected by all the climate and weather models used for forecasting, is the effect of ocean waves on the polar sea ice covers. When waves travel through sea ice they can cause it to break, and the extent to which ice is broken up can effect the rate at which it melts and freezes, and its ability to drift with the current. Over the summer I looked at the first model to be developed that attempts to predict the extent of ice breakage by sea ice.
The Waves in Ice Model (WIM) is the product of decades of experimental and theoretical research by academics from all around the world (Williams et al. 2013). This model is still in its earliest stages, it is only capable of making predictions along a one dimensional cross section of the ice cover.
Often parameters in models need to be estimated. For instance, when modelling sea ice it is difficult to decide how elastic the ice should be, as sea ice can be very different in different parts of the ocean. In WIM, we estimate a value for the elasticity of the ice, then test to see how sensitive the model is to variations of this value. If there is little sensitivity, then we can justify our estimation, at least until more accurate data becomes available. The focus of my summer research was to conduct a number of such sensitivity tests on the WIM.
Over the next year I will be completing an honours year for my mathematical sciences degree, and aim to improve WIM by including a second spatial dimension. This will allow the model to be coupled with both wave and ice models, and hopefully become included in broader climate and weather forecasting models.
IPCC (2007), `Fourth assessment report’. http://www.ipcc.ch/publications_and_data/ar4/syr/en/spms3.html
Williams, T. D., Bennetts, L. G., Squire, V. A., Dumont, D. & Bertino, L. (2013), `Wave-ice interactions in the marginal ice zone. part 1: Theoretical foundations’, Ocean Modelling (Submitted) pp. 1-43.