Numerical simulations using 1-D Lagrangian cloud model and 2-D cloud model with spectral (bin) microphysics show that large CCN (natural of seeded) decrease the height of raindrop formation. As a result, raindrops fall through the cloud or in the vicinity of cloud where the air humidity is high. The raindrops do not experience any significant evaporation, so the lost in precipitating mass is minimum. In the absence of large CCN, raindrops (or frozen particles) form at higher levels and often fall down outside of clouds through the air with low humidity. As a resulIt, a significant fraction of precipitation mass can be lost. Thus, the efficiency of seeding depends of cloud characteristics, as well as on the thermodynamical and dynamical properties of environmental flow.

Simulations with the Lagrangian model shows that there is an optimum size of seed particles (about 2.5 microns) that provides the formation of maximum rainwater mass at a given mass of a seed reagent. Factors affecting this size will be discussed. We will present also the results of hygroscopic seeding of clouds using also 2-D cloud spectral microphysics model.