Abstract:The accuracy of precipitation monitoring carries significant importance and impact on weather forecasting, climate change, service decision-making, disaster prevention and mitigation. The scientific and rational layout of the precipitation station network directly affects the precision of precipitation monitoring. Selecting Ya’an, Sichuan Province as the research object, based on the existing station network layout, we are analysing the control range of stations, the distance and the precipitation correlation between the closest stations, and on the principle of “no loss of precipitation information”. We are proposing an optimisation method for station network layout based on the reduction in stations and evaluating the rationality of station network layout by the changes in area rainfall before and after the reduction. The results show that there are 265 precipitation stations in Ya’an, with a control area of 2.64-510.78 km2 for each station, a distance between the nearest stations of 0.3-12.28 km, and a correlation coefficient of precipitation between the nearest stations of 0.62-0.993. Stations with smaller control areas and shorter distances between the nearest stations have relatively higher correlation coefficients of precipitation between the nearest stations. After each reduction, the control area of precipitation stations and the distance between the nearest stations gradually increase, and the precipitation correlation coefficient between the nearest stations gradually decreases. From the mutual information amount and the distribution of stations, the layout of the station network is becoming more uniform and reasonable. The monthly and daily area rainfall before and after reduction do not linearly increase with the increase of station reduction times, but the changes are not significant. The area rainfall and variation amplitude of the Tyson polygon method are both smaller than those of the arithmetic mean method. The area rainfall after the previous reduction aligns with that of the original network, with a smaller error. From the six precipitation processes in Ya’an, the heavy rainstorms of the six precipitation processes are mainly concentrated in the central part of Ya’an. After each reduction, the rain area and range are essentially consistent with the original station network, and the rain centre and region of heavy rain are able to be captured, as well as the extreme value of precipitation. However, with the reduction of stations, the granularity of precipitation morphology description has declined. Overall, after each reduction, the station network maintains a good ability to detect and capture precipitation, indicating that the optimization method for station network layout proposed in this article is reasonable.