China will make use of more advanced technologies to improve weather forecasts and provide a broader array of customized services, the nation's meteorological administration said in a white paper published on Saturday.
With 10,596 ground monitoring stations and 198 radar facilities, Chinese meteorologists have become increasingly accurate in predicting the weather, the paper said.
It said the accuracy of 24-hour forecasts for thunderstorms and heavy rainfall last year rose by 4.9 and 8.5 percentage points, respectively, compared with the average rates between 2014 and 2016.
The National Meteorological Center issued more than 210,000 weather alerts in 2017 with increased speed and efficiency that helped those affected better prepare for coming hazards, the white paper said.
The average effective lead time needed for the center to issue alerts before major meteorological events fell from 10 minutes to between 5 and 8 minutes last year, it said.
With its meteorological-geological disaster forecasting and warning system, the administration offers daily reports on the probability of the occurrence of disasters such as mountain torrents and debris flow. This allowed the meteorological administration to spare people additional danger from some 1,016 geological disasters last year, and may have helped avoid as many as 40,000 casualties, it said.
The administration has offered forecasting services that can be updated every three hours for the country's expressway network, and forecasts were even applied to areas as small as 25 square kilometers beginning last year.
It has also published a 2018-20 proposal to offer smart forecasting services for the country's entire transportation network.
Administration spokesman Chen Zhenlin, said China has increasingly been turning to the internet of things and artificial intelligence to broaden its capabilities in meteorological data collection and weather forecast accuracy. One example is a monitoring device as small as a badge that can be attached to clothing and luggage.
"With the badge, people can help collect meteorological data anytime and anywhere. This could assist quite a bit in increasing the density of observation points and further improving the accuracy of forecasts. The larger the database is, the more accurate the forecast will be," he said, noting that the technology has already been used in some regions.
Shen Yun, a weather official, said the administration has been cooperating with Tsinghua and Nanjing universities to apply artificial intelligence in automatic weather forecasting.
"The current cooperation mainly focuses on immediate forecasts. For longer-term forecasts, there still needs to be more cooperation with algorithm engineers and experts with knowledge of weather and meteorology," she said.