Abstract
Abstract. The properties of the multiple-scale instabilities present in a non-hydrostatic forecast model are investigated. The model simulates intense convection episodes occurring in northern Italy. A breeding technique is used to construct ensembles of perturbations of the model trajectories aimed at representing the instabilities that are responsible for error growth on various timescales and space scales. By means of perfect model twin experiments it is found that, for initial errors of the order of present-day analysis error, a non-negligible fraction of the forecast error can be explained by a bred vector ensemble of reasonable size representing the growth of errors on intermediate scales. In contrast, when the initial error is much smaller, the spectrum of bred vectors representing the fast convective-scale instabilities becomes flat, and the number of ensemble members needed to explain even a small fraction of the forecast error becomes extremely large. The conclusion is that as the analysis error is decreased, it becomes more and more computationally demanding to construct an ensemble that can describe the high-dimensional subspace of convective instabilities and that can thus be potentially useful for controlling the error growth.
Highlights
In the last 10–15 years there has been operational interest in non-hydrostatic, convection-resolving models (e.g. Dixon et al, 2009; Seity et al, 2011; Baldauf et al, 2011) and in the possibility of making use of data assimilation (e.g. Zhang et al, 2003; Kain et al, 2010; Schenkman et al, 2011; Claussnitzer et al, 2011) to improve short-range forecasting and nowcasting of weather fields, or the accuracy of a past event trajectory reconstruction
The error during the first 10 h grows much faster in the small initial error trajectory (R21), whereas, during the second episode, the error growth of the two trajectories is very similar. This is because the growth rate is dominated by fast instabilities when the initial error is small, but, as the forecast error grows in time, these components saturate and slower instabilities associated with larger dynamical scales become dominant3
We use the model as a laboratory where we assume that a model trajectory represents the truth and we can vary the amplitude and structure of the initial condition error
Summary
In the last 10–15 years there has been operational interest in non-hydrostatic, convection-resolving models (e.g. Dixon et al, 2009; Seity et al, 2011; Baldauf et al, 2011) and in the possibility of making use of data assimilation (e.g. Zhang et al, 2003; Kain et al, 2010; Schenkman et al, 2011; Claussnitzer et al, 2011) to improve short-range forecasting and nowcasting of weather fields, or the accuracy of a past event trajectory reconstruction. In the last 10–15 years there has been operational interest in non-hydrostatic, convection-resolving models Prediction and data assimilation systems for the ocean and the atmosphere are largely founded upon the theory, developed over the last decades, of predictability and state estimation in chaotic dynamical systems. Atmospheric NWP (numerical weather prediction) models that predict the synopticscale evolution of mid-latitude weather systems are based on the hydrostatic assumption and include parameterisation of convective processes. Ensemble forecasting and data assimilation for such models have reached a mature stage, whereas non-hydrostatic models face the hard task of dealing with the fast instabilities typical of convection. A comparison of the timescales involved in prediction of hydrostatic versus non-hydrostatic models has been performed by Hohenegger and Schär (2007a), casting some doubts on the feasibility of borrowing algorithms developed in the context of the former for the application to the latter models
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
