Plans and tests for the operational implementation of the new Canadian Land Data Assimilation System (CaLDAS)

A new Canadian Land Data Assimilation System (CaLDAS) is currently being developed and tested in order to improve land surface initial conditions in numerical prediction systems that will become operational at Environment Canada (EC) in 2011 or 2012. The core component of this new system is an external land surface model driven by best estimates of atmospheric forcing and of geophysical land surface characteristics. Various CaLDAS configurations are currently being tested and compared. They include versions using Ensemble Kalman Filter or simple variational techniques (and eventually an hybrid of these two approaches). Furthermore, assimilation of screen-level data (typically used in national weather centers to initialize soil moisture and surface temperatures) and of space based remote sensing data for soil moisture and terrestrial snow is being evaluated. Results supporting EC's new strategy for land data assimilation will be presented at the conference, along with a longer term view of CaLDAS applications in air quality, climate, and hydrology (i.e., versions including carbon, vegetation, ground water, …).