Study: AI Predicts Immediate Rainfall Better than Existing Systems
2021-10-21
LRC
TXT
大字
小字
滚动
全页
1British researchers say they have created an artificial intelligence (AI) model that is highly effective at predicting rainfall within the next 90 minutes.
2The model was built by scientists at Google-owned research company DeepMind in London.
3The company partnered with researchers at the University of Exeter in Britain and the Met Office, the British government's national weather service.
4The team says tests of the system showed it produced more accurate predictions, or forecasts, for short-term rainfall than other existing systems.
5The paper recently appeared in the publication Nature.
6The scientists centered on a kind of weather prediction known as "precipitation nowcasting."
7This is defined as the prediction of precipitation, like rain, within the next 1-2 hours.
8Making forecasts of this kind has traditionally been very difficult.
9But accurate nowcasting can be a very important tool for preparing for possible weather emergencies.
10It can be used by officials in the areas of water management, agriculture, air and ground traffic control and outdoor event planning.
11The Met Office has depended on a system called numerical weather prediction since the early 1950s.
12That method uses "the latest weather observations alongside a mathematical computer model of the atmosphere" to make forecasts.
13The system collects data on the level and movements of fluid in the atmosphere based on radar and satellite data.
14The computer then processes the data and produces map-like models to predict future weather.
15The Met Office currently produces forecasts with a powerful new "supercomputer" that became fully operational in 2017.
16The agency says the increased power permits forecasters to make more accurate predictions based on detailed models.
17But the supercomputer is designed to produce forecasts for very wide areas over a period of hours, days or a week.
18The computer still struggles with precipitation predictions within 1 to 2 hours.
19The DeepMind team says its model uses machine learning to capture the amount, timing and place of rainfall within 90 minutes to 2 hours.
20Machine learning involves feeding massive amounts of data to computers to train the machines to learn patterns and behaviors over time.
21In this case, the system was trained to make predictions of future radar based on past radar readings.
22The study used radar data from both Britain and the United States.
23"We were especially interested in the ability of these models to make predictions on medium to heavy-rain events, which are the events that most impact people and the economy," the team said.
24The scientists said their model aims to create predictive "movies" of immediate weather patterns.
25They said that, unlike other systems, theirs can accurately capture large-scale events, as well as several different rain possibilities.
26This combination permits "rainfall uncertainties" to be deeply explored.
27During the testing process, the team said the model showed "statistically significant" improvements in these areas compared to competing methods.
28In addition, the researchers reported that a group of more than 50 weather experts from the Met Office rated the AI model to be more accurate and useful in 89 percent of cases than existing forecasting systems.
29Despite the study results, researchers admitted that the model still has limitations.
30They said "more work is needed to improve the accuracy of long-term predictions and the accuracy on rare and intense events."
31The team said it will now center on developing new methods to test and measure the model's performance.
32It will also work to create more specialized systems to work in real-world conditions.
33I'm Bryan Lynn.
1British researchers say they have created an artificial intelligence (AI) model that is highly effective at predicting rainfall within the next 90 minutes. 2The model was built by scientists at Google-owned research company DeepMind in London. The company partnered with researchers at the University of Exeter in Britain and the Met Office, the British government's national weather service. 3The team says tests of the system showed it produced more accurate predictions, or forecasts, for short-term rainfall than other existing systems. The paper recently appeared in the publication Nature. 4The scientists centered on a kind of weather prediction known as "precipitation nowcasting." This is defined as the prediction of precipitation, like rain, within the next 1-2 hours. Making forecasts of this kind has traditionally been very difficult. 5But accurate nowcasting can be a very important tool for preparing for possible weather emergencies. It can be used by officials in the areas of water management, agriculture, air and ground traffic control and outdoor event planning. 6The Met Office has depended on a system called numerical weather prediction since the early 1950s. That method uses "the latest weather observations alongside a mathematical computer model of the atmosphere" to make forecasts. 7The system collects data on the level and movements of fluid in the atmosphere based on radar and satellite data. The computer then processes the data and produces map-like models to predict future weather. 8The Met Office currently produces forecasts with a powerful new "supercomputer" that became fully operational in 2017. The agency says the increased power permits forecasters to make more accurate predictions based on detailed models. 9But the supercomputer is designed to produce forecasts for very wide areas over a period of hours, days or a week. The computer still struggles with precipitation predictions within 1 to 2 hours. 10The DeepMind team says its model uses machine learning to capture the amount, timing and place of rainfall within 90 minutes to 2 hours. 11Machine learning involves feeding massive amounts of data to computers to train the machines to learn patterns and behaviors over time. In this case, the system was trained to make predictions of future radar based on past radar readings. The study used radar data from both Britain and the United States. 12"We were especially interested in the ability of these models to make predictions on medium to heavy-rain events, which are the events that most impact people and the economy," the team said. 13The scientists said their model aims to create predictive "movies" of immediate weather patterns. They said that, unlike other systems, theirs can accurately capture large-scale events, as well as several different rain possibilities. This combination permits "rainfall uncertainties" to be deeply explored. 14During the testing process, the team said the model showed "statistically significant" improvements in these areas compared to competing methods. In addition, the researchers reported that a group of more than 50 weather experts from the Met Office rated the AI model to be more accurate and useful in 89 percent of cases than existing forecasting systems. 15Despite the study results, researchers admitted that the model still has limitations. They said "more work is needed to improve the accuracy of long-term predictions and the accuracy on rare and intense events." 16The team said it will now center on developing new methods to test and measure the model's performance. It will also work to create more specialized systems to work in real-world conditions. 17I'm Bryan Lynn. 18Bryan Lynn wrote this story for VOA Learning English, based on reports from DeepMind, the Met Office and Nature. Susan Shand was the editor. 19____________________________________________ 20Words in This Story 21artificial intelligence (AI) - n. an area of computer science that deals with giving machines the ability to seem like they have human intelligence 22accurate - adj. correct or exact 23management - n. the way in which something is carried out and controlled 24pattern - n. a particular way that something is often done or repeated 25impact - n. an effect that a person, event or situation has on someone or something 26uncertain - adj. not sure or not able to decide about something 27statistic - n. a fact in the form of a number that shows information about something 28significant - adj. important or notable