外刊 | The high-tech race to improve weather forecasting, Part 3

The high-tech race to improve weather forecasting, Part 3

    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。

    第三部分介绍私人预测机构的优势与功能。

Private prognostications 私人预测

1.天气预测的功能

The WMO reckons numerical forecasting will approach that theoretical limit sometime around 2050. But that leaves plenty of room for improvement in the meantime.

The ECMWF presently produces accurate forecasts of daily weather—meaning it can predict things like the temperature and when it will rain, give or take a couple of degrees or hours—around the globe at least a week ahead of time.

It has, on occasion, successfully predicted certain big events, like hurricanes, up to ten days ahead.

that theoretical limit 前文提到的数值预测限制

2.引出私人预测

But big global or regional forecasts are not the only game in town. There is also a growing demand for faster or more specific forecasts than can be provided by public institutions (which, being mostly funded by taxpayers, tend to produce what will be the most helpful to the most people).

Private companies are filling the gaps.

the only game in town 最重要的东西

fill the gap 填补空白

3.以IBM为例介绍私人预测机构

In 2016, for instance, IBM, an American computing firm, bought the Weather Company, which specialised in combining different governmental models, for an estimated $2bn. (Sceptics joked that IBM had invested in the wrong type of cloud.) 

Within a year the firm began selling “hyper-local” forecasts to businesses, designed to predict the weather in a small area between two and 12 hours ahead. By 2020, according to Comscore, an American media-analytics firm, IBM was the biggest provider of weather forecasts in the world.

4.私人机构的优势：自由度高，数据处理少，可实时预报

The firm’s success stems, in part, from its freedom to pick its own priorities. Predicting the weather only a few hours ahead drastically reduces the amount of number-crunching required. 

That, says Peter Neilley, the Weather Company’s chief meteorologist, allowed the firm to develop a global model with a 3km resolution that churns out a new forecast once an hour. (The ECMWF’s high-resolution global model, by contrast, produces a new forecast every six hours.)

stem from 来自

priority 优先事项

number-crunching 数据处理

churns out 炮制，粗制滥造，大量产出

5.私人机构的另一优势：可以付费使用公共机构的数据，却不必向公共机构共享数据

Alongside its own model, the Weather Company still sucks in the output of publicly funded forecasters around the world. That reveals another private-sector perk. 

Some national and international agencies, including both the Met Office in Britain and the ECMWF, can charge businesses that use their output. But all are obliged to make them available. The pipeline does not have to flow in the other direction.

suck in 收集

perk 好处

6.功能细分：精确时间

In recent years, private offerings have become even more specific. Companies are increasingly aware of how the weather affects their work.

For instance, wind and solar energy producers—and the electricity grids to which they are connected—rely on knowing what the weather will do in the next few hours. Other applications are less obvious.

Deliveroo, a food-delivery firm, knows that it must account for the effect of rain on traffic when working out the fastest way to transport a pad Thai from one side of a city to another.

private offering 这里指私人预测

Deliveroo 谐音像kangaroo但不是美团外卖

a pad Thai 泰式炒粉

7.功能细分：自选信息

Meteomatics, a Swiss firm founded in 2012, allows its customers to crunch data from a range of sources in a way that suits their needs—such as “downscaling” the output of a numerical model by shaping it around the local topography. 

Those customers, say Alexander Stauch and Rob Hutchinson, two of the firm’s senior managers, increasingly want to pipe that data directly into their own algorithms. 

Energy traders, for example, predict gas prices based on how much wind or sunshine is around to generate wind or solar power.

downscale 缩小比例

topography 地形地貌

pipe the data 导入数据

be around to =be able to 可以（前者强调条件，后者强调能力）

8.功能细分：自选地区

Meteomatics（前文提到的瑞士公司） also aims to fill in gaps in observational data for places their clients are interested in. To that end, it flies its own fleet of sensor-covered drones.

In May Tomorrow.io, an American firm founded in 2015, began launching satellites that are likewise designed to help plug data holes around the world. Its main product, though, is “weather intelligence” software that turns forecasts into instructions.

The Bill and Melinda Gates Foundation, one of the world’s biggest charities, uses the company to send text messages to farmers in sub-Saharan Africa, advising them on when best to plant their crops.

end 目的

drone 无人机

fill in gaps =plug the holes 填补空缺

9.私人机构帮助贫穷国家获取天气预测信息

Private players insist their participation is beneficial for everyone. There are far more weather stations in rich countries than poor ones. 

“Outside of America, western Europe, Japan and Australia, and a couple of other countries, national meteorological services are lagging decades behind,” says Rei Goffer, one of Tomorrow.io’s founders.

Some rich-country agencies help other countries—the Met Office, for example, works with the governments of India, South Africa and several South-East Asian countries.

Even so, Mr Goffer argues, many countries simply cannot afford the sort of good-quality forecasting that might help them adapt to a changing climate. Tomorrow.io’s satellites aim to allow countries access to better weather infrastructure without having to build it from scratch.

build it from scratch= start all over 从零开始

欢迎指正，感谢阅读。