“轉自：燈塔大數據；微信：DTbigdata”
————————————————————————————————————————————————————————————
天氣狀況與人類息息相關，不管你身處在地球的哪個角落，都無法忽視天氣和氣候給人類生活造成的影響。全球各個地區的氣候有着巨大的差異，然而生活在某個固定區域的人可能不會明顯感覺到這種差異。
天氣和氣候狀況也是影響一個地區經濟的重要因素，氣候一旦出現異常，經濟就可能會受到嚴重的負面影響，比如洪澇災害會損壞重要的基礎設施，而長期不下雨則會導致乾旱的發生。
氣候對社會來說影響巨大。有人甚至認爲敘利亞內戰就是由天氣變化所導致的，美國前任國務卿約翰·克里說過：“氣候變化不一定是敘利亞危機的直接導火索，但是很明顯，敘利亞發生的旱災是火上澆油、雪上加霜。”
因此，關注天氣狀況並不是僅僅爲了知道明天應該穿什麼，而是需要公司和企業制定一個長期的計劃，根據天氣數據判斷未來幾周的天氣狀況會對其業務造成什麼樣的影響。
衆所周知，英國的連鎖超市樂購（Tesco）就利用氣候數據預測銷售情況，調整倉庫庫存。舉例來說，如果他們接到天氣預報數據說接下來三週會有明顯升溫，他們就會檢查倉庫中是否有足夠的烤肉架、防曬霜和冷飲等來滿足消費者的需求。
樂購2013年的數據就顯示，當年憑藉天氣預測數據，該集團節約了超過600萬英鎊，將脫銷成本降低了30%。
但是2013年之後的幾年，全球氣候變得更加變幻莫測，2014年、2015年、2016年是有記錄以來全球氣溫最高的幾個年份，給人們預測天氣增加了很多不確定因素的干擾，給公司營銷方案的制定造成了很大困擾。
難以捉摸的氣候造成的負面影響包括：損壞基礎設施從而導致供應鏈中斷、惡劣天氣導致農作物收成差，從而導致某些產品供應短缺。
從這種意義上來說，氣候數據不僅僅能夠幫助人們預測災害的發生，還能幫助人們制定合理的方案，以減少突發氣候災害對公司業務所造成的負面影響。
影響氣候的數據點數量龐大，想要挨個輸入這些數據點，然後推測出某個地區出現不同天氣得可能性，這無疑是個異常艱鉅的任務，需要消耗無數的人力和能源。
美國國家海洋和大氣局（NOAA）投入巨資打造克雷（Cray）超級計算機，這種超級計算機的計算速度高達每秒3千萬億次。
要將如此多影響氣候變化的因素考慮在內，就必須依靠這種強大的計算力，越來越多的公司和企業也將會依賴於這些氣候數據來幫助他們提高經營的效率。
衆所周知，在公司經營中，外部影響因素和內部主觀能動同樣重要，天氣和氣候就是外部因素的其中一方面。天氣和氣候對每個公司都有一定程度的影響，因此我們有必要好好的利用氣候數據來預測天氣狀況，並制定出高效的應對方案。
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.
翻譯：燈塔大數據
