Scholarlink

<p>In this work, we integrate high-resolution measurements of air pollutants with a transformer-based masked autoencoder to explore fine-scale spatial relationships among pollutants, local meteorology, and land cover across a heterogeneous urban area. Using a custom-built miniaturized broadband cavity-enhanced spectrometer (mBBCEAS) for NO<sub>2</sub>, along with PM<sub>1</sub>, PM<sub>2.5</sub>, O<sub>3</sub>, and meteorological sensors, we conducted 66 mobile surveys across the 1.1 km<sup>2</sup> study area over three seasons. A transformer-based masked autoencoder, pretrained on synthetic data, accurately reconstructed full pollutant and meteorological fields from heavily masked, multi-variable observations (<i>R</i>² = 0.89) and precisely classified concentration and meteorological intensity levels into ten quantile-based categories (F1 = 92.9% with one-bin tolerance). Attention-derived feature relevancy revealed fine-scale transport and identified key drivers, including winds and land cover, that strongly modulate ground-level pollutant gradients over tens of meters. The results further demonstrate the feasibility of optimized, data-driven urban air-quality sampling strategies.</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

Learning neighborhood-scale cross-dependencies among air pollutants, meteorology and land cover using mobile sensing and transformers

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年3月6日

查看全文
  • D. Nissenbaum,
  • S. Bagon,
  • R. Sarafian,
  • C. C. Womack,
  • O. Sapir,
  • S. S. Brown,
  • Y. Rudich

摘要

In this work, we integrate high-resolution measurements of air pollutants with a transformer-based masked autoencoder to explore fine-scale spatial relationships among pollutants, local meteorology, and land cover across a heterogeneous urban area. Using a custom-built miniaturized broadband cavity-enhanced spectrometer (mBBCEAS) for NO2, along with PM1, PM2.5, O3, and meteorological sensors, we conducted 66 mobile surveys across the 1.1 km2 study area over three seasons. A transformer-based masked autoencoder, pretrained on synthetic data, accurately reconstructed full pollutant and meteorological fields from heavily masked, multi-variable observations (R² = 0.89) and precisely classified concentration and meteorological intensity levels into ten quantile-based categories (F1 = 92.9% with one-bin tolerance). Attention-derived feature relevancy revealed fine-scale transport and identified key drivers, including winds and land cover, that strongly modulate ground-level pollutant gradients over tens of meters. The results further demonstrate the feasibility of optimized, data-driven urban air-quality sampling strategies.