The CSIRO-QCCCE contribution to CMIP5 using the CSIRO Mk3.6 climate model

Abstract

Fifth phase of the Climate Model Intercomparison Project (CMIP5) is the principal framework for coordinated climate modeling experimentation supporting the preparation of the IPCC 5 Assessment Report to be released in 2013. About 20 modeling groups from around the world are undertaking the CMIP5 experiments and model data is being hosted on the Earth System Grid which consists of international data nodes and gateways (see http://cmip-pcmdi.llnl. gov/cmip5/). The CMIP5 experimental design features a standard set of model simulations consisting of simulations based on historical concentration and emission changes, simulations along potential future pathways to 2100, and a range of idealised experiments to understand climate sensitivity and to quantify key feedbacks. Projections of future climate are based on the new generation of emission scenarios, which were developed by the research community (Moss et. al., 2010). The CMIP5 experimental protocol provides four emission scenarios based on the Representative Concentration Pathways (RCPs), which are identified with their amounts of net radiative forcing input into the Earth's climate system at the end of the 21 Century. They include following scenarios:• A no policy RCP8.5 W/m ; • Two stabilization RCP6 W/m and RCP4.5 W/m ; • A peak and decline RCP2.6 W/m . CSIRO Marine and Atmospheric Research and the Queensland Climate Change Centre of Excellence (QCCCE) are contributing to this international project using the CSIRO-Mk3.6 Atmosphere Ocean Global Climate Model (AOGCM). The Mk3.6 climate model is a substantial upgrade from its recent predecessors Mk3.0 and Mk3.5, which were used to contribute to CMIP3. Upgrades included addition of an interactive aerosol scheme, which explicitly treats sulfate, dust, sea salt and carbonaceous aerosol. It also includes an updated radiation scheme and other changes to the atmospheric physics component. The CSIRO Mk3.6 climate model has been used to perform most of the long-term simulations which include the core experiments plus a number of tier1 and tier2 experiments specified by the CMIP5 experimental protocol (Taylor et al., 2011a). This paper shows selected results from historical simulations for the period 1851 to 2005 using various combinations of radiative forcing (both natural and anthropogenic) and results from climate change projections for the period 2006 to 2100. The analysis quantifies the relative contribution of various radiative forcing factors to the simulated changes in mean annual surface temperature during the 1981 to 2005 period. Results from the analysis show that simulations with natural forcing only resulted in small changes in the surface temperature, whereas the simulations with all forcing show statistically significant changes in the surface temperature over most of the Earth surface and these changes have magnitude comparable to observed changes. Furthermore, the rate of warming is greatest in the experiment with greenhouse gases only, where the cooling impact of anthropogenic aerosols is not accounted for. Model simulations show continuing warming and drying in subtropical land areas during the early part of 21 century.