The value of target observations

This project, realized with G Kelly and J-N Thepaut of the Satellite Section and Carla Cardinali of the Data Assimilation Section, aims to address two key questions:

(i) What is the value of ocean data for forecasts verified downstream over land? For example, what is the value of Pacific ocean data for short-range forecasts verified over North America, or for medium-range forecasts verified over Europe? And what is the value of Atlantic ocean data for short-range forecasts verified over Europe?

(ii) What is the value of targeted data taken over ocean in small regions identified using singular vector methods? Are data taken in these SV-based regions more valuable than data taken in similar-size, but randomly chosen regions?

These questions have been addressed using a set of OSSE-type data depletion and/or data thinning experiments.

The project started in January 2005. A pdf file with some key results (August 2005, taken from a lecture given at NASA/GMAO) can be accessed from this page.

• The value of ocean and targeted observations

(site map)