The session aims to discuss ways and tools to extract relevant and useful information in big data sets produced from regional climate modelling. The philosophy behind downscaling is to make the best use of available relevant information about the local climate, inferred from dependencies to large-scale conditions and local geography. Observations provide important information, but projections must also account for the range of outcomes associated with natural variability.
In addition, users of regional and local climate projections face a growing problem as monitoring, global modeling, and regional statistical and dynamical modelling efforts increase the already vast volumes of data accumulated.
The character of the data differ depending on their nature, the modelling strategy and model set-up (e.g. parameterisation schemes). Some of the results are difficult to compare, such as downscaling results derived from the use of regional climate models and empirical-statistical downscaling.
There are, however, statistical methods designed to extract relevant information from data that have not yet been fully made use of, there are statisticians who are interested in climate analysis (e.g. Thorarinsdottir, et al., 2014; Chanddler, 2013; Knutti et al., 2010).
-Thorarinsdottir, Thordis, Jana Sillmann, and Rasmus Benestad (2014). “Studying Statistical Methodology in Climate Research.” Eos, Transactions American Geophysical Union 95, no. 15: 129–129. doi:10.1002/2014EO150008.
-Chandler R.E. (2013). Exploiting strength, discounting weakness: combining information from multiple climate simulators. Phil Trans R Soc A 371: 20120388, doi: 10.1098/rsta.2012.0388.
-R Knutti, R Furrer, C Tebaldi, J Cermak, GA Meehl(2019 Challenges in combining projections from multiple climate models
Journal of Climate 23 (10), 2739-2758