Scholarlink

<p>Air pollution is reckoned as one of the main critical factors in risks related to human health as well as to the environment, especially in key areas like the Arctic, where global warming is affecting the ecosystem at an alarming rate. Major pollutants, such as Particulate Matter (PM<sub>10</sub>), are thus crucial to monitor for efficient planning ahead of extreme events: to that extent, a reliable short to medium-term forecast is vital for prompt action. We propose a Deep Learning approach to air pollution forecasting aimed at predicting 48-h concentrations of PM<sub>10</sub>. We analyse the behaviour of said pollutant together with auxiliary factors, such as meteorological variables and outputs from state-of-the-art numerical models, through several forefront time-series-specialised architectures, i.e. Transformers. We develop an enhanced version of the best-performing Transformer architecture to define a suitable model to forecast PM<sub>10</sub> concentrations from a wide variety of stations across Northern Europe. Our proposed methodology manages to outperform pivotal numerical models and associated postprocessing, offering a valuable alternative for local air pollution forecasting.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A transformer approach to forecasting PM10 concentration in the Arctic and Northern Europe

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年5月6日

查看全文
  • Alice Cuzzucoli,
  • Ilaria Crotti,
  • Srdjan Dobricic,
  • Antonello Pasini

摘要

Air pollution is reckoned as one of the main critical factors in risks related to human health as well as to the environment, especially in key areas like the Arctic, where global warming is affecting the ecosystem at an alarming rate. Major pollutants, such as Particulate Matter (PM10), are thus crucial to monitor for efficient planning ahead of extreme events: to that extent, a reliable short to medium-term forecast is vital for prompt action. We propose a Deep Learning approach to air pollution forecasting aimed at predicting 48-h concentrations of PM10. We analyse the behaviour of said pollutant together with auxiliary factors, such as meteorological variables and outputs from state-of-the-art numerical models, through several forefront time-series-specialised architectures, i.e. Transformers. We develop an enhanced version of the best-performing Transformer architecture to define a suitable model to forecast PM10 concentrations from a wide variety of stations across Northern Europe. Our proposed methodology manages to outperform pivotal numerical models and associated postprocessing, offering a valuable alternative for local air pollution forecasting.