外刊 | The high-tech race to improve weather forecasting, Part 2

The high-tech race to improve weather forecasting, Part 2 

            这篇文章有点长，我把它分成了考研阅读的长度（600-700words）。

            第一部分介绍了欧洲中期天气预报中心的功能、极端天气与天气预报的重要性。

                            在结尾用1960s至今预测准确率的提升引出第二部分。

            第二部分介绍了天气预测的原理及预测模型计算问题。

Cloud computing 云彩计算

1.天气预测的原理及其依赖计算机性能的原因

Most of that improvement has been down to more powerful computers, says Tim Palmer, a meteorologist and physicist at the University of Oxford. Weather forecasts work by carving the world into a grid of three-dimensional boxes. 

Each is populated with temperature, air pressure, wind speed and the like, and the system’s evolution simulated by grinding through enormous numbers of calculations.

（1）be down to =depend on取决于

（2）carve (something) into (something) 把……刻进

（3）"grid boxes"在气象学中是指将地球表面划分成的三维网格，用于气象模型。每个格子代表一个空间区域，后文中9km、16km都是“格子”的不同规格。后文提及的"resolution"（分辨率）是指气象模型中用于划分网格的大小的级别。格子越小，分辨率越高，数据越精细，模型越精确。

（4）is populated with（=is full of）

        “populate”及物动词，意为“居住”。

（5）“simulated”是过去分词形式。

        当你想随便加点相关信息的时候，就可以用分词短语的结构。

（6）grind “研磨，加工”，可以翻译为“处理”

2.模型越精确效果越真实

Better computers allow finer models. In the same way that a high-resolution digital photo looks more realistic than a coarse-grained one, using a smaller grid helps match a model more closely to the real world. 

The ECMWF’s highest-resolution global model, for instance, chops the globe into boxes that are 9km square, down from 16km in 2016, and splits the atmosphere vertically into more than 100 layers.

high-resolution 高分辨率的

coarse-grained 粗糙的

3.小网格的作用

Smaller grids also allow models to recreate more of what happens in the real weather. “Deep convective clouds”, for instance, are formed as hot air floats upwards. They can produce heavy rain, hail and even tornadoes, but typically cannot be resolved with grids bigger than about 5km.

Models have instead represented them using stopgap code that acts as a simplified substitution.

convective cloud 对流云
stopgap 应急的

stopgap code指编程中不够理想的代码

4.精确度与局域性，有钱才能全都要

But smaller grids come at a high price. Halving the horizontal size of a grid means that four times as many boxes—and four times as many calculations—are needed to cover a given area. One option is to trade resolution for locality. 

The sharpest offering from the National Oceanic and Atmospheric Administration, in America, for instance, uses grid boxes 3km square, but covers only North America. Computing, meanwhile, continues to improve. 

The world’s fastest computer is Frontier, installed at Oak Ridge National Laboratory in Tennessee. Using it, ECMWF scientists were able to experiment with running a worldwide model with a 1km resolution.

halve 把……分成两半

the sharpest in America 美国最精确的模型

5.混沌系统之限，无关技术能力

But no matter how powerful computers become, there is a limit to how far ahead a numerical forecast can look. The atmosphere is what mathematicians call a “chaotic system”—one that is exquisitely sensitive to its starting conditions. 

A tiny initial change in temperature or pressure can compound over days into drastically different sorts of weather. Since no measurement can be perfectly accurate, this is a problem that no amount of computing power can solve.

In 2019 American and European scientists found that even the most minor alterations to simulations resulted in highly divergent forecasts for day-to-day weather after about 15 days. 

“It seems to be a limit that nature sets,” explains Falko Judt, a meteorologist at the National Centre for Atmospheric Research, in America. “It has nothing to do with our technological capabilities.”

"Starting conditions"（初始条件）是指在数值模型（numerical forecast）中用来启动模拟或预测的初始状态，非常依赖准确的数据。

欢迎指正，感谢阅读。