面部識(shí)別技術(shù)可以預(yù)測極端天氣

科學(xué)家稱，面部識(shí)別技術(shù)可以預(yù)測極端天氣

Machine learning technology used in facial recognition could predict extreme weather events responsible for billions of dollars of damage every year, according to a new study published in the Monthly Weather Review.

根據(jù)發(fā)表在《天氣評(píng)論》(weather Review)月刊上的一項(xiàng)新研究，用于面部識(shí)別的機(jī)器學(xué)習(xí)技術(shù)可以預(yù)測每年造成數(shù)十億美元損失的極端天氣事件。

Storms that produce hail can have large, damaging impacts on agriculture, property, and even wildlife. Just last week, as many as 13,000 shorebirds and waterfowl were killed in a severe hailstorm in eastern Montana. Such storm events cost as much as $22 billion every year in damages to people and property, according to CBS News. More than 4,600 major hailstorms occurred in 2018, according to NOAA’s Severe Storms database, with the majority of damages being reported in the central part of the country.

產(chǎn)生冰雹的風(fēng)暴可能對農(nóng)業(yè)、財(cái)產(chǎn)甚至野生動(dòng)物造成巨大的破壞性影響。就在上周，在蒙大拿州東部一場嚴(yán)重的冰雹中，多達(dá)13000只濱鳥和水禽死亡。據(jù)哥倫比亞廣播公司新聞報(bào)道，這類風(fēng)暴事件每年造成的人身和財(cái)產(chǎn)損失高達(dá)220億美元。根據(jù)國家海洋和大氣管理局的嚴(yán)重風(fēng)暴數(shù)據(jù)庫顯示，2018年發(fā)生了4600多場大冰雹，大部分損失報(bào)告在美國中部地區(qū)。

But the size and severity of hailstorms are often difficult to predict. That’s where artificial intelligence technology by the National Center for Atmospheric Research (NCAR) comes into play. Rather than zooming in on the features of a face, scientists have trained a deep learning model called “convolution neural network” to pinpoint specific storm features in order to determine whether hail will be formed and, if so, how large the hailstones will be.

但冰雹的大小和嚴(yán)重程度往往難以預(yù)測。這就是國家大氣研究中心(NCAR)的人工智能技術(shù)發(fā)揮作用的地方。科學(xué)家們訓(xùn)練了一種叫做“卷積神經(jīng)網(wǎng)絡(luò)”的深度學(xué)習(xí)模型，來精確定位特定的風(fēng)暴特征，以確定冰雹是否會(huì)形成，如果會(huì)形成，那么冰雹會(huì)有多大。

"We know that the structure of a storm affects whether the storm can produce hail," said NCAR scientist David John Gagne in a statement. "A supercell is more likely to produce hail than a squall line, for example. But most hail forecasting methods just look at a small slice of the storm and can't distinguish the broader form and structure."

NCAR科學(xué)家大衛(wèi)·約翰·加涅在一份聲明中說：“我們知道，風(fēng)暴的結(jié)構(gòu)會(huì)影響風(fēng)暴是否會(huì)產(chǎn)生冰雹。”例如，超級(jí)單體比暴風(fēng)線更容易產(chǎn)生冰雹。但是，大多數(shù)冰雹預(yù)報(bào)方法只關(guān)注風(fēng)暴的一小部分，無法區(qū)分更廣泛的形式和結(jié)構(gòu)。”

A perfect recipe of meteorological ingredients allows for a storm to produce hailstones, but even when conditions are ripe, the size and severity of hailstones will vary depending on the path and conditions within the storm, which is collectively known as the “storm structure”.

氣象成分的完美配方才可以促進(jìn)風(fēng)暴產(chǎn)生冰雹，但即使條件成熟，冰雹的大小和嚴(yán)重程度也會(huì)因風(fēng)暴的路徑和條件而變化，這就是所謂的“風(fēng)暴結(jié)構(gòu)”。

"The shape of the storm is really important," Gagne said. "In the past, we have tended to focus on single points in a storm or vertical profiles, but the horizontal structure is also really important."

“風(fēng)暴的形狀非常重要，”加涅說。“過去，我們傾向于關(guān)注風(fēng)暴的單點(diǎn)或垂直剖面，但其實(shí)水平結(jié)構(gòu)也非常重要。”

NCAR scientists presented the machine learning software with images of simulated storms paired with data about temperature, pressure, and wind speed and direction, along with simulations of hail based on those factors. The program then figured out which features correlated with whether or not it hails and how big the hailstones are.

NCAR的科學(xué)家向機(jī)器學(xué)習(xí)軟件展示了模擬風(fēng)暴的圖像，以及溫度、壓力、風(fēng)速和風(fēng)向的數(shù)據(jù)，以及基于這些因素的冰雹模擬。然后，該程序自動(dòng)計(jì)算出哪些特征與冰雹發(fā)生、以及冰雹大小相關(guān)。

Generally speaking, the model confirmed storm features that the team had previously linked to hailstones. However, it’s important to note a number of limitations, including the fact that simulated storms vary dramatically from actual storms. Regardless, the team says their research could eventually transition into operational use to potentially replace the complex mathematical predictions currently used.

總的來說，該模型證實(shí)了該團(tuán)隊(duì)此前與冰雹有關(guān)的風(fēng)暴特征。然而，需要注意的是一些限制，包括模擬風(fēng)暴與實(shí)際風(fēng)暴之間的巨大差異。無論如何，研究小組表示，他們的研究最終可能會(huì)轉(zhuǎn)化為可操作性的應(yīng)用，有可能取代目前使用的復(fù)雜數(shù)學(xué)預(yù)測。