AccuWeather is modernising weather forecasting by using Microsoft Azure Databricks and other Azure services to build a forecasting engine that converts traditional weather data into a format for artificial intelligence (AI)-assisted analysis.
Weather forecasting uses data in a format called General Regularly distributed Information in Binary form (GRIB 2). This format cannot be read by many modern data processing platforms, but AccuWeather wanted to filter its data through AI, machine learning and other digital technologies to maintain the reliability and timeliness of its weather forecasts.
“We pride ourselves on superior accuracy in our forecasts and making use of the latest information available is critical – weather can’t wait,” said Scott Mackaro, vice president of science and innovation at AccuWeather. “We wanted to push the boundaries of our capabilities and create faster, more accurate, and more localised forecasts for our customers.”
AccuWeather began using Azure Databricks to support its next-generation forecasting engine and convert GRIB data into formats for AI-assisted analysis on Azure.
Now, AccuWeather staff can easily interact with datasets and ask specific questions about weather parameters such as temperature and precipitation in specific locations at specific times. They can then use other Azure services to apply machine learning models to the company’s datasets and train predictive models to improve the speed and accuracy of AccuWeather forecasts.
“With some types of severe weather forecasts, it can be a life-or-death scenario,” said Christopher Patti, chief technology officer at AccuWeather. “With Azure, we’re agile enough to process and deliver severe weather warnings rapidly and offer customers more time to respond, which is important when seconds count and lives are on the line.”