• English 
  • Spanish 

SPECS-PREFACE workshop on initial shock, drift and systematic error

A workshop took place at IC3 with the objective of discussing the recent developments in initial shock, drift and systematic error assessment in a climate prediction context. This workshop was motivated by the fact that, despite the large communities working in the identification of the reasons for the large range of model biases, very little work has been done in understanding the causes of the initial shock and in ways to reduce the impact of the drift. Scientists involved in the European projects SPECS and PREFACE were involved in this workshop whose format consisted in a set of brief presentations, followed by several discussion sessions. The questions addressed included the identification of the physical processes responsible for the model drift and the initial shock, their characterization, suggestions for model improvements that reduce the drift building on the continuous efforts to reduce the systematic error, considerations of how the initialisation strategy influences the skill and the best way to deal with the drift and the systematic error a posteriori and the need for bias correction. During the discussion, the attendants concluded that work should be done to understand how the initialisation affects the simulated variability, to distinguish between initial shock and drift, to investigate new methods to perform bias correction, to distinguish between the stationary and non stationary components of the error, to assess the impact of the initial shock on the skill, and to consider the sensitivity of the three terms to the parameter and model uncertainty. This talk will illustrate some of these aspects and offer more detailed conclusions and plans for the future in several European institutions.