Abstract: In order to improve the accuracy of radar quantitative precipitation retrievals,combining Doppler multi-radar mosaic data and the gauge-observed hourly rainfall intensity data,the radar quantitative precipitation retrieval scheme was optimized in Liaoning province based on the established Z-I relationship using the optimization method.The results show that the evaluation indexes of the two optimization schemes for quantitative precipitation retrieval are improved.Overall,the two optimization schemes reduce the errors of radar quantitative precipitation retrieval effectively.The capability of precipitation retrieval is further improved.The precipitation retrieval results for Fushun “8.16” typical strong precipitation indicate that the spatial distribution of the radar quantitative precipitation retrieval using the two optimization retrieval schemes showed high accuracy compared to the surface precipitation data.In particular,the problems that the precipitation zone above 20.0 mm·h^-1 is smaller than actually happening and the magnitude of the short-time strong precipitation above 40.0 mm·h^-1 is underestimated are largely corrected.After using the optimization scheme,all the evaluation indexes are significantly improved and the uncertainties and dispersions in the radar quantitative precipitation retrieval are smaller than before.Precipitation retrieval results have better stability and reliability,which provide a quantitative reference for short-term forecasting and early warning of heavy rainfall events.