Dealing with uncertainty: how numerical weather prediction simulators work, a talk by Dr Michel Mesquita (M2Lab.org and SKL, Norway)
Abstract
Numerical weather prediction models, or simulators, have been widely used for weather forecasting and climate prediction in the past decades. Recent improvements in the dynamic core and parameterisation schemes have made it possible to improve the quality of such predictions.
Still there are uncertainties depending on the model configuration: initial conditions, choice of schemes, coupling, nesting, and scenario, among others. Hence, the focus of this talk will be on limited area models (LAMs), their mathematical treatment and the different sources of uncertainty.
LAMs rely on data from global circulation models to calculate values for the more detailed weather variables at a shorter time scale with higher resolution grid points within the boundaries. Examples will be provided based on a LAM called the Weather Research and Forecasting (WRF) model.
WRF makes it possible to create very high resolution simulations for weather / climate purposes by nested domains, idealised studies, and large-eddy simulations. Best practices on dealing with uncertainty in numerical weather prediction will be discussed throughout, based on the WRF lectures on m2lab.org.
This talk will be given remotely, via Zoom. If you wish to register for attendance at this London branch talk, please contact Gemma Reeves who will provide the required details.
No charge is made to attend meetings. Non-IMA members are welcome.
