气候数据只能预测降雨么？那你就错了！

“转自：灯塔大数据；微信：DTbigdata”

————————————————————————————————————————————————————————————

天气状况与人类息息相关，不管你身处在地球的哪个角落，都无法忽视天气和气候给人类生活造成的影响。全球各个地区的气候有着巨大的差异，然而生活在某个固定区域的人可能不会明显感觉到这种差异。

天气和气候状况也是影响一个地区经济的重要因素，气候一旦出现异常，经济就可能会受到严重的负面影响，比如洪涝灾害会损坏重要的基础设施，而长期不下雨则会导致干旱的发生。

气候对社会来说影响巨大。有人甚至认为叙利亚内战就是由天气变化所导致的，美国前任国务卿约翰·克里说过：“气候变化不一定是叙利亚危机的直接导火索，但是很明显，叙利亚发生的旱灾是火上浇油、雪上加霜。”

因此，关注天气状况并不是仅仅为了知道明天应该穿什么，而是需要公司和企业制定一个长期的计划，根据天气数据判断未来几周的天气状况会对其业务造成什么样的影响。

众所周知，英国的连锁超市乐购（Tesco）就利用气候数据预测销售情况，调整仓库库存。举例来说，如果他们接到天气预报数据说接下来三周会有明显升温，他们就会检查仓库中是否有足够的烤肉架、防晒霜和冷饮等来满足消费者的需求。

乐购2013年的数据就显示，当年凭借天气预测数据，该集团节约了超过600万英镑，将脱销成本降低了30%。

但是2013年之后的几年，全球气候变得更加变幻莫测，2014年、2015年、2016年是有记录以来全球气温最高的几个年份，给人们预测天气增加了很多不确定因素的干扰，给公司营销方案的制定造成了很大困扰。

难以捉摸的气候造成的负面影响包括：损坏基础设施从而导致供应链中断、恶劣天气导致农作物收成差，从而导致某些产品供应短缺。

从这种意义上来说，气候数据不仅仅能够帮助人们预测灾害的发生，还能帮助人们制定合理的方案，以减少突发气候灾害对公司业务所造成的负面影响。

影响气候的数据点数量庞大，想要挨个输入这些数据点，然后推测出某个地区出现不同天气得可能性，这无疑是个异常艰巨的任务，需要消耗无数的人力和能源。

美国国家海洋和大气局（NOAA）投入巨资打造克雷（Cray）超级计算机，这种超级计算机的计算速度高达每秒3千万亿次。

要将如此多影响气候变化的因素考虑在内，就必须依靠这种强大的计算力，越来越多的公司和企业也将会依赖于这些气候数据来帮助他们提高经营的效率。

众所周知，在公司经营中，外部影响因素和内部主观能动同样重要，天气和气候就是外部因素的其中一方面。天气和气候对每个公司都有一定程度的影响，因此我们有必要好好的利用气候数据来预测天气状况，并制定出高效的应对方案。

 

英文原文

Weather data isn’t just about predicting rain

The weather is perhaps the one universal that every human can relate to, it is the universal connector between everyone regardless of where you are from. Most people have seen broadly the same kinds of weather, despite geographical differences meaning that others see more extreme versions than others.

It is also one of the most important elements of any economy, after all, with any kind of excess of one type of weather, there is likely to be significant disruption, whether this is excessive precipitation destroying vital infrastructure or over abundance of dry weather causing droughts. It also has a profound impact on society as a whole, with some even saying that weather has played a part in starting the Syrian civil war, with John Kerry, then Secretary of State saying ‘I’m not telling you that the crisis in Syria was caused by climate change, but the devastating drought clearly made a bad situation a lot worse.’

It is therefore not simply a case of looking at the weather forecast to ascertain what to wear that day, companies need to be able to use long-term planning to predict how the weather in several weeks may impact them.

UK based supermarket chain Tesco have become well known for their use of weather data to help predict sales and stock requirements. For instance, if they can predict that there will be a heatwave in 3 weeks time, they can then make sure that they have increased stocks of disposable barbecues, sun cream and cold drinks to cater for the increased demand. When this was widely reported in 2013 it was said that the chain had managed to save £6m ($7.5m) per year and seen a cutting out-of-stock by 30% on special offers.

However, since 2013, when Tesco’s use of this data became more publicly known, the world’s weather has become considerably more complex. 2014, 2015, and 2016 have all been the hottest years on record, which has caused more unpredictability in our weather and more disruption to companies. This can be anything from supply chains being disrupted due to infrastructure damage through to shortage of stock because farmers couldn’t grow a specific crop due to difficult conditions. It means that the use of data is essential not only in the prediction of these events, but in the planning in case they occur.

There are a huge and diverse number of data points that can impact changes in the weather and trying to input every one and then predict with any degree of certainty down to localities is incredibly difficult and requires considerable power. It is why the NOAA invested millions in a Cray supercomputer that processes 3 quadrillion calculations per second. The huge variations in conditions that go into weather conditions requires this kind of power and is something that more and more companies are needing to look at to help run their businesses effectively.

Everybody knows that running a business effectively is as much about awareness of external influences as it is about what you do internally, and weather is the ultimate example of that. Every company can be impacted and the use of data to predict and react to it is essential.

 

翻译：灯塔大数据