Modern methods for spatial detailing (downscaling) of meteorological fields with an insufficient spatial resolution are considered. The basic advantages and disadvantages of statistical, physically based and dynamical-statistical approaches are briefly discussed. The greater attention is paid to the dynamical methods based on high resolution atmosphere models. Examples of works in which different downscaling techniques were used are listed. A conclusion: all downscaling methods possess advantages and disadvantages, a researcher needs to choose the best method proceeding from the specific problem to solve.
The nesting of high resolution atmosphere model into general circulation model grid is considered in short. Various ways methods of keeping simulations close to coarse data are analyzed. A conclusion: a method of spectral nudging is the best for downscaling purposes. The spectral nudging is easier for implementing in spectral models.
Comparison of intermittent incremental data assimilation and dynamical downscaling is spent. A conclusion: these techniques are close and some blocks of one technique can be used in the other.
Nowadays there are freely available high resolution atmosphere models (in particular, WRF ARW, RSM), that contains nudging techniques. These models are dynamical downscaling systems almost ready to use.
Key words: Downscaling, dynamical downscaling, atmosphere modeling, climate modeling, nudging dynamic model state to a given state.