The Monash Simple Climate Model Experiments (MSCM-DB v1.0): An interactive database of mean climate, climate change and scenario simulations

Dietmar Dommenget, Kerry Nice, Tobias Bayr, Dieter Kasang, Christian Stassen, Michael Rezny
2018 Geoscientific Model Development Discussions  
<p><strong>Abstract.</strong> This study introduces the Monash Simple Climate Model (MSCM) experiment database. The model simulations are based on the Globally Resolved Energy Balance (GREB) model. They provide a basis to study three different aspects of climate model simulations: (1) understanding the processes that control the mean climate, (2) the response of the climate to a doubling of the CO<sub>2</sub> concentration, and (3) scenarios of external <i>CO<sub>2</sub></i> concentration and
more » ... lar radiation forcings. A series of sensitivity experiments in which elements of the climate system are turned off in various combinations are used to address (1) and (2). This database currently provides more than 1,300 experiments and has an online web interface for fast analysis of the experiments and for open access to the data. We briefly outline the design of all experiments, give a discussion of some results, and put the findings into the context of previously published results from similar experiments. We briefly discuss the quality and limitations of the MSCM experiments and also give an outlook on possible further developments. The GREB model simulation of the mean climate processes is quite realistic, but does have uncertainties in the order of 20&amp;ndash;30<span class="thinspace"></span>%. The GREB model without flux corrections has a root mean square error in mean state of about 10<span class="thinspace"></span>°C, which is larger than those of general circulation models (2<span class="thinspace"></span>°C). However, the MSCM experiments show good agreement to previously published studies. Although GREB is a very simple model, it delivers good first-order estimates, is very fast, highly accessible, and can be used to quickly try many different sensitivity experiments or scenarios.</p>
doi:10.5194/gmd-2018-143 fatcat:inpchj5t7ne4nlmk5fsyfr6p4q