Fifty years of numerical modeling of baroclinic ocean |
| |
| Authors: | A S Sarkisyan |
| |
| Institution: | 1.Institute of Numerical Mathematics,Russian Academy of Sciences,Moscow,Russia |
| |
| Abstract: | This paper presents a brief critical analysis of the main historical stages of numerical modeling for the last fifty years. It was a half a century ago that the numerical simulation of an actual baroclinic ocean was initiated by the author and his students 1, 2]. In meteorology, studies on the numerical modeling of a baroclinic atmosphere existed much earlier 21, 22]. Despite this, a similar move in oceanography was met with strong resistance. At that time, there were many studies on the calculation of the total mass transport. The founders of this field, V.B. Shtokman, H. Sverdrup, and W. Munk, were mistaken in believing that they addressed baroclinic models of the ocean. The author preferred works by V. Ekman 12] and I. Sandström and B. Helland-Hansen 19]. A generalization of recent studies made it possible to come to some conclusions on the need to use the level of the free oceanic surface as a basis rather than the function of total mass transport, on the role of the baroclinic β effect (BARBE), on the joint effect of baroclinicity and bottom relief (JEBAR), etc. The author conditionally divides these fifty years into the following three stages. (1) The first stage was 1961–1969, when the author and his students performed almost exclusively diagnostic and adaptation calculations of climatic characteristics. (2) The second stage began with papers by K. Bryan 23] and his students. This is an important and promising stage involving mainly prognostic studies and four-dimensional analysis. The major advances in modeling at this stage (the Gulf Stream separation point 61], the Kuroshio seasonal evolution 63], the formation of the cold intermediate layer in the Black Sea 80], the subsurface countercurrent in the Caspian Sea 25], the realistic four-dimensional analysis of the Kara Sea 60], etc.) were due to high-resolution and/or data assimilation with an adequate period of integration. (3) The third stage began with the activities of international intercalibration programs such as the Arctic Ocean Model Intercomparison Project (AOMIP), the Global Ocean Data Assimilation Experiment (GODAE), Coordinated Ocean-Ice Reference Experiments (COREs), etc. Despite some defects initially, this is the most significant stage. For example, there is still very little data on GODAE, and COREs data are often used for a comparison of integral characteristics, the reliability of which cannot be established by direct measurements. |
| |
| Keywords: | |
| 本文献已被 SpringerLink 等数据库收录！ |
|
