Modelling and simulation are an increasingly important part of modern science, especially in policy-relevant disciplines such as weather and climate. Good practice in the design, use and interpretation of models and their output is therefore vital, both for sound science and for informing evidence-based policy decisions.
I’ll discuss some of the methods used to extract information from models about the future climate and related phenomena that is of value to decision-makers. This involves developing robust and fair forecast evaluation techniques, as well as combining information from different sources, such as from ensembles of simulation models, from empirical models based on the statistics of past observations, and from an understanding of the underlying physical processes within the climate system, all of which contain some uncertainty.
Methods to quantify the capabilities (and limitations) of our models and their uncertainties allow a better understanding of the spatial and temporal scales at which simulation models can provide useful (quantitative) predictive information. They can also provide insight about when the use of simulation models, in combination with other forms of information, can lead to better decision-making in the context of climate variability and change.
Dr Emma Suckling is a researcher in climate sciences at the University of Reading, having recently moved from the London School of Economics. She has a PhD in theoretical physics from the University of Surrey.