NOAA research shows meteorologists continue to outperform algorithms when forecasting the weather, particularly in adverse conditions
their ability to observe and draw connections where algorithms cannot—gives these forecasters an edge that continues to outperform the glitzy weather machines in the highest-stake situations.” https://www.wired.com/... Rob Perillo / @robperillo : “When it comes to forecasting the elements, many seem ready to welcome the machine. But humans still outperform the algorithms—especially in bad conditions.” https://www.wired.com/...
A good human forecaster - as @shawnmilrad alludes to in the article - will be able to interpret and contextualize both the raw and post-processed forecast. They can then _add value_ to different stakeholders or consumers. (3/10)
We should be skeptical of folks hawking AI as a “solution” to weather (and climate) in general - especially given that the field already has a rich tradition of developing quite advanced ML-based applications. (7/10)
Interesting short article on @WIRED this morning - https://www.wired.com/... but I don't think it really captures the challenge, opportunity, and pitfalls here. A short thread. (1/10)
Weather apps have come a long way, but they still can't beat human forecasters, particularly when the weather is anything but sunny and dry. https://www.wired.com/... @CTVStanfield, @AdrianaYZhang, @ryanhardingctv will always beat the forecast on your phone.
Data-driven modeling and analysis is already widely employed in the weather world, even if it's not always super-sexy deep learning. No one would ever offer raw NWP model output as a “forecast” - we'd defer to statistically post-processed products like a MOS or the NBM (2/10)
Humans continue to outperform weather models: “Experience—their ability to observe and draw connections where algorithms cannot—gives these forecasters an edge that continues to outperform the glitzy weather machines in the highest-stake situations.” https://www.wired.com/...
The crux here is that AI has to integrate synergistically within existing meteorological products, processes, and value chains. AI provides tools which might be useful - but for the right problems and to the right stakeholders/decision makers. (6/10)
“When it comes to forecasting the elements, many seem ready to welcome the machine. But humans still outperform the algorithms—especially in bad conditions.” https://www.wired.com/...
Your automated weather forecast might seem like magic, but human beings are still essential for predicting the inclement stuff (and won't be replaced by computers anytime soon). Also, we need to protect our NWS. https://www.wired.com/...
“If we want to continue to receive in-depth weather forecasts and crucial warnings, touched by human hands, we need to preserve agencies and services that value human-augmented forecasts...” The Danger of Leaving Weather Prediction to AI | WIRED https://www.wired.com/...