Fig. 1. Spatial distribution map of ground rain gauge

Fig. 2. Flow chart of precipitation recognition algorithm based on random forest

Fig. 3. The importance ranking of precipitation identification variables based on random forest

Fig. 4. Parameter sensitivity analysis of precipitation recognition model based on random forest

Fig. 5. Comparison of the precipitation zone identification results of this study with those of the GSMaP_NOW

Fig. 6. Comparison of the precipitation zone identification results of this study with those of the FY4A QPE

Fig. 7. Comparison of model or product’s daily cumulative precipitation area on July 8， 2019 （UTC time） with observation results of surface rain gauge

Fig. 8. The change of hourly daily precipitation recognition and evaluation index of the algorithm in July 2019

Fig. 9. The algorithm in this study scored hourly skills in July 2019 （the box chart represents the 25th， 50th and 75th percentiles， and the outside of the box extends to 1.5 times of the quartile difference （75th percentile minus 25th percentile）， and the outliers are represented by hollow circle）

Fig. 10. This research model and GSMaP_NOW product and FY4A QPE product hourly precipitation identification and evaluation index time series broken （0：00， July 8， 2019~22：00， July 9， 2019， UTC time ）