AKIO ARAKAWA HAS been a leader in the field of atmospheric general circulation model (AGCM)
development from its beginning. AGCMs are essential tools for studies of global warming and projecting the consequences of anthropogenic climate change. Arakawa's inventiveness and extraordinary insight on atmospheric processes have resulted in fundamental contributions to the design of AGCMs in several areas, primarily: 1) numerical schemes suitable for the long model integrations required by climate studies; 2) modeling of cloud processes including cumulus parameterization; and 3) modeling of planetary boundary layer (PBL) processes. His work has influenced the design of practically all numerical weather and climate prediction models.
Arakawa obtained his B.Sc. in physics and D.Sc. in meteorology from the University of Tokyo in 1950 and 1961, respectively. From 1961 to 1963 he visited the University of California Los Angeles (UCLA) and worked with Yale Mintz on developing the Mintz-Arakawa AGCM, the first among several generations of what would be the UCLA AGCM. Versions in each generation (currently the VII) have been made available to other institutions for further development and application. In 1965, Arakawa joined the faculty at UCLA, where he is currently Professor Emeritus and Research Professor.
In AGCMs the governing equations of fluid motion are written in a format suitable for numerical integration with a high-speed computer. The format used is particularly important in long-term integrations, since inadequacies can result in distortion of the solutions and unrealistic amplifications of kinetic energy. Arakawa designed elegant numerical schemes in which such unrealistic amplifications cannot occur and distortions were reduced. The schemes guarantee that flow properties constrained to be either constant or bounded in the continuous equations remain so in the corresponding expressions numerically solved by the computer. In this way, he derived the Arakawa Jacobian in late 1961, which was followed over the years by the derivation of families of numerical schemes for AGCMs.
The most widely influential of Arakawa's works in the context of global warming and climate change has been the parameterization for AGCMs of the effect of clouds generated by convection. The parameterization problem consists of formulating the collective effect of processes not resolved by the model grid in terms of the resolvable-scale prognostic variables, and is a crucial part of general circulation modeling. Arakawa struggled with the cumulus parameterization problem for several years. The breakthrough came in collaboration with his graduate student Wayne Schubert during the early 1970s. The basic tenet in the Arakawa-Schubert parameterization is the approximate cancellation (quasi equilibrium) between destabi-lization of the atmosphere by large-scale processes and stabilization by convection. Arakawa and several of his students would later refine the original formulation. The parameterization is still used today in different versions, including the Relaxed and Simplified Arakawa-Schubert schemes (RAS and SAS, respectively).
Arakawa has also made major contributions to the parameterization of PBL processes. These determine the complex interactions between the atmosphere and Earth's surface and as such must be represented correctly in climate models. PBL processes, however, are dominated by turbulent transport and mixing, which AGCMs in general cannot explicitly resolve. Furthermore, the PBL can be topped by stratiform clouds, which are also difficult to explicitly resolve. Errors at the surface due to misrepresentations of PBL cloud effects can spuriously amplify due to feedbacks, particularly over the ocean.
For the parameterization of PBL processes, Arakawa adopted a unique framework based on including a variable depth PBL in the model, which becomes the lowest layer in the UCLA AGCM generation IV. The PBL depth is explicitly predicted using the mass budget relationship, in which mass at the top can be added through turbulence processes and removed by cumulus mass flux for given large-scale conditions above the PBL. The latter term can be provided by the Arakawa-Schubert cumulus convection parameterization. In this way, interactions between PBL and cumulus convection processes can be modeled.
The increased power of computers has challenged the unambiguous separation between resolvable processes, which can be highly transient, and parameterized unresolvable processes, which can only be near a statistical equilibrium. Arakawa is currently working on the unification of those formulations and the reduction of the artificial dependency of model physics on grid size.
sEE ALso: Climate Models; Climatic Data, Atmospheric Observations; Global Warming.
BIBHoGRAPHY. Akio Arakawa, General Circulation Model Development: Past, Present and Future, Chapters 1 and 23 (D. A. Randall. Ed., Academic Press, 2000). Carlos R. Mechoso and Akio Arakawa, General Circulation Models, Encyclopedia of Atmospheric Sciences (J. R. Holton, J. Pyle, and J. A. Curry, Eds. (Academic Press, 2003). Akio Arakawa, The Cumulus Parameterization Problem: Past, Present, and Future (J. Climate, 2004).
C. R. Mechoso University of California, Los Angeles
Was this article helpful?