Data Assimilation Research in Support of Ross Island Meteorological Experiment (RIME)

University Corporation For Atmospheric Res, Boulder CO

Investigators

Abstract

This award will provide funds to develop and implement an advanced data assimilation system for use in a regional numerical weather forecasting model (MM5) specifically designed for the Ross Sea and McMurdo area of Antarctica. With the lack of traditional observations over Antarctica and the Southern Ocean, and with the difficulty in significantly augmenting traditional observations, the optimal use of satellite and other non-traditional, e.g. automatic weather station, data is currently the only viable path to improve regional weather analysis and prediction. Further, the lack of data also hampers progress in meteorological research and model development, since there is insufficient data to study physical processes and to represent them accurately in numerical models. Three tasks are included in this project: (1) development of a mesoscale variational data assimilation system over the Antarctic region; (2) assimilation of automatic weather station and satellite data and assessment of their impact on short-range prediction over the Antarctic region; and (3) performing forecast sensitivity experiments with the MM5 adjoint model to study what types of data, over what regions, are most important to accurate forecast over the Ross Sea area. The project will also form one of the bases of a planned Ross Island Meteorological Experiment.