The predictionmodel for the highestand lowest temperature is established by the CMSVM regressionmethod based on ECMWF data and the extreme temperature observations from 8 automaticweather stations inQingyang ofGansu Province from 2003 to2007.The effect test in 2008 operation forecast indicated that the rate ofaverage forecastaccuracy is64% and 71% for the highestand lowest temperature in 5 days, which has the good instruction for the real-timeweather service. The forecast accuracy rate of the highest and lowest temperature decreased with the forecast time increase, and the accuracy rate for the lowest temperature is higher than thatof the highest temperature. The forecast effect for the lowest temperature is better in spring and summer than that in autume and winter, but it is opposite for the highest temperature. The absolute and the average forecasterrors for the highestand lowest temperature are added with the forecast time increase, and the absolute forecasterror for the highest temperature is bigger than thatof the lowestbut the average error is close and positive. The forecast effect test in Septemper of 2008 indicated that the forecast values presented the similar trend to real time observations for the highest and lowest temperature.

Based on the routinemeteorological data and the Dopp ler radar date, the causes of a hail p rocess occurred on August 2, 2006 in Qingyang were analyzed from the synop tic environment, physical quantity field and the characteristics ofDopp ler radar echoes. Results show that itwas a typ ical northwest flow pattern at the level of 500 hPa, the lower air convergence and the upper air divergence,the coup ling of the positive vortex in themiddle troposphere led to the liftingmovement develop ing in the low level, and thiswas advantageous to the upward transportation of the local water vapor; the short wave trough at the north of Hetao area caused the cold air to move down and this became the main triggeringmechanism of this p rocess. The Dopp ler radar data analysis showed that the p rocess occurred after a strong convection cloud which created by a squall line, and there was a good corresponding relationship between the change of the cloud top temperature and the development of the convection cloud; the merge of the single cloud cluster was advantageous to the energy enhancement and easy to p roduce the hail.

The calibrated analysis of the p recip itation estimation by radar was made using Kalman filter method based on p recip itation data measured by rain - gaug in two rainfall p rocesses occurred in July and Sep tember of 2007 in Qingyang of Gansu Province, and the errors were also analyzed by comparing the measured data by rain - gauge. The results show that the effect of calibration in the p recip i2 tation p rocess from 03: 00 to 11: 00 July 28 surpasses that in rainfall p rocess from 14: 00 to 22: 00 Sep tember 6, and the relative error dropped from 166. 4% to 7. 6% and 284. 3% to 56. 9% after calibration, respectively. The random noises in the p rocess of rainfall es2 timation by radar can be effectively eliminated and the p recision of quantitative estimation can be enhanced by the Kalman filtermethod for the regional p recip itation, meanwhile, the fine structure of p recip itation exp loration by radar can be retained better.