Transient-induced Climate Drift

Ake Johansson Department of Meteorology, Stockholm University, Arrhenius Laboratory, Stockholm, Sweden

Search for other papers by Ake Johansson in
Current site
Google Scholar
PubMed
Close
,
Brian Reinhold Department of Meteorology, Stockholm University, Arrhenius Laboratory, Stockholm, Sweden

Search for other papers by Brian Reinhold in
Current site
Google Scholar
PubMed
Close
, and
Suranjana Saha Department of Meteorology, Stockholm University, Arrhenius Laboratory, Stockholm, Sweden

Search for other papers by Suranjana Saha in
Current site
Google Scholar
PubMed
Close
Restricted access

Abstract

The climate drift of various quantities associated with deep, planetary-scale, equilibrated, transient Rossby waves are estimated for the Southern Hemisphere extratropical summer as revealed by the DERF II (Dynamical Extended Range Forecasting) dataset. It is found that the vertical structures of these waves systematically become too baroclinic during the course of integration. There are two time scales associated with this climate drift. There is one very short time scale, estimated to be of the order of one day, when the waves become more barotropic. It is followed by a period when the wave baroclinicity monotonically increases, and after roughly 10 days the model structures appear to have reached their statistically equilibrated state.

In the meantime, the kinetic energy of the transient waves decreases substantially to roughly half the observed value. After this initial drop, however, the transient kinetic energy increases again, and it is not clear if an equilibrium value has been reached after 30 days, which is the limit of the DERF II dataset. This third time scale is not found in the quantities directly associated with the vertical structures per se, but it is hypothesized to be a consequence of these errors.

A theory is utilized that in a simplified way takes into account the processes that determine the vertical structure of baroclinic waves as well as their robustness as a means of understanding the processes leading to these errors. The implications from this theory are that the formulation and magnitude of the dissipative and diffusive processes in the model are the most likely problem, but there are other possibilities.

Abstract

The climate drift of various quantities associated with deep, planetary-scale, equilibrated, transient Rossby waves are estimated for the Southern Hemisphere extratropical summer as revealed by the DERF II (Dynamical Extended Range Forecasting) dataset. It is found that the vertical structures of these waves systematically become too baroclinic during the course of integration. There are two time scales associated with this climate drift. There is one very short time scale, estimated to be of the order of one day, when the waves become more barotropic. It is followed by a period when the wave baroclinicity monotonically increases, and after roughly 10 days the model structures appear to have reached their statistically equilibrated state.

In the meantime, the kinetic energy of the transient waves decreases substantially to roughly half the observed value. After this initial drop, however, the transient kinetic energy increases again, and it is not clear if an equilibrium value has been reached after 30 days, which is the limit of the DERF II dataset. This third time scale is not found in the quantities directly associated with the vertical structures per se, but it is hypothesized to be a consequence of these errors.

A theory is utilized that in a simplified way takes into account the processes that determine the vertical structure of baroclinic waves as well as their robustness as a means of understanding the processes leading to these errors. The implications from this theory are that the formulation and magnitude of the dissipative and diffusive processes in the model are the most likely problem, but there are other possibilities.

Save