Abstract: 
In middle tendays and last tendays of February in 2006, there were tow preci
pitation processes in Changjiang delta, i.e. snow and rain respectively, but the
 forecast of precipitation type was wrong. A comparison of these two processed s
hows that in the snow process, there was west wind in the front of trough and th
e circulation was flat, whereas in the rain process, there was southwest wind in
 the front of trough.  Snow would happen only if the cold advection, which was r
esulted from the northeast return current, generated a cold center over east Chi
na.  However the north and northwest wind decreased the air temperature, but no 
precipitation would happen at the same time, so there would be no snow in this case.  Before snowing the air 
at 850 hPa was dry and before raining the air at lowmiddle levels was wet.  Th
e temperatures at 700~800 hPa and in the inversion layer were important to the 
precipitation type.  At the end of snow process the air in high level become dry
 first, whereas at the end of rain process the air in middle level become dry fi
st.    In snow process the snow and ice crystals were centralized in the high le
vel, and the center of snow content was at 600~700 hPa.  In rain process, the l
ower boundary of snow content lifted up and its center was in a level above 600 
hPa.  The center of snow content weakened very fast when it moved down northward
s.