abstract |
|
Data assimilation is a procedure of production of an accurate "image" of the
true state of the atmosphere at a given time. This image is used then as the
initial state (or input parameters) for a numerical weather prediction (NWP)
system. Basically data assimilation process may be divided into three parts:
- gathering of measurements obtained by various observation systems;
- reduction of the observation data sets (data thinning);
- combination of the observations with an a priory estimate of the model
state, i.e. previous wheather forecast, to produce a most accurate numerical
description of the atmosphere.
The main goal of this talk is to give a brief overview of the the assimilation
process concentrating on the 1st and 3d parts. The main kinds of observation
systems and several most important assimilation techniques will be presented.
With respect to the data thinning (the 2d part) only the necessity of such
data reduction will be demonstrated, whereas data thinning algorithms are
reserved as a subject of subsequent talks.
|