Met Office

Dr Christine Johnson

Christine is researching ensemble forecasts for application in data assimilation.

About

Christine is researching ensemble forecasts for application in data assimilation.
Christine is researching ensemble forecasts for application in data assimilation.

Areas of expertise:

  • Ensemble forecasting
  • 4D-Var
  • ETKF
  • Multi-model ensemble combination
  • Forecast calibration

Publications 

Current activities

Christine's research mainly focuses on the generation of ensemble forecasts, for use in hybrid data assimilation methods. Her work couples together the strengths of the Met Office in both short-range ensemble forecasting and variational data assimilation.

The main purpose of ensemble forecasting is to determine the most likely outcome of the weather based on a set of equally-likely model forecasts. A new application is to use the ensemble as a proxy for the uncertainty in the background state (prior) used in 4D-Var. One example is to use the ensemble to generate a static climatological error covariance of the background state. A second example is to use the ensemble to provide flow-dependent covariances that can be combined with a static climatological covariance derived from another source. In both these applications, it is essential that the ensemble forecasts provide a good proxy for the forecast error. The aim of Christine's work is to examine the properties of the ensemble forecasts and make improvements for the benefit of hybrid data assimilation.

Christine's research also focuses on the assessment of multi-model ensembles for medium-range weather forecasting, as part of the international research collaboration known as THORPEX. This includes how to calibrate and combine forecasts produced from different operational centres. Research has shown that by combining ensemble forecasts from different models, we are able to get a better representation of the errors resulting from the forecast model, and therefore giving better probability forecasts.

Career background

Christine first joined the Met Office in 2006 working on multi-model ensembles as part of THORPEX. In 2010, her research changed focus to the generation of ensemble forecasts for use in data assimilation.

Prior to joining the Met Office, Christine completed a BSc in Mathematics and Meteorology and a PhD on 4D-Var, both at the Department of Meteorology at the University of Reading. She then went on to spend two years as an ASP post-doctoral fellow at NCAR, working on adaptive observations.

Last Updated: 21 April 2011