ACCElERAtINg PROgRESS IN glOBAl AtMOSPhERIC MODEl DEVElOPMENt thROUgh IMPROVED PARAMEtERIzAtIONS

G lobal models are used in many applications, ranging from Numerical Weather Prediction (NWP) to seasonal prediction and climate simulation. Despite great computational advances, calculations using these models can still only afford grid spacing for which many of the important processes occurring in the atmosphere—most notably boundary layer processes as well as moist convection and cloud processes—remain unresolved. It is therefore necessary to represent those subgrid-scale processes as a function of the grid-scale variables. The technique to achieve this representation is usually referred to as parameterization (Stensrud 2007). Despite impressive progress in the development of parameterizations over the last few decades, solutions to some long-standing model problems remain elusive. Examples for such problems are the erroneous representation of the ITCZ and the associated cold sea surface temperature (SST) bias in the central tropical Pacific (e.g., Zhang et al. 2007), the poor representation of the Madden–Julian oscillation and other modes of tropical variability (e.g., Lin et al. 2006), the incorrect representation of the frequency of occurrence of highand low-intensity rainfall events (e.g., Sun et al. 2006), the poor representation of the diurnal cycle of rainfall (e.g., Yang and Slingo 2001; Betts and Jakob 2002), as well as ongoing difficulties in simulating the El Niño–Southern Oscillation (ENSO) phenomenon (e.g., Neale et al. 2008). It is noteworthy that many of these shortcomings are ascribed to the poor representation of subgrid-scale processes, in ACCElERAtINg PROgRESS IN glOBAl AtMOSPhERIC MODEl DEVElOPMENt thROUgh IMPROVED PARAMEtERIzAtIONS Challenges, Opportunities, and Strategies