Scientific Publications

NextAIRE Researchers Share New Study on AI-Based Prediction of Atmospheric Particle Size Distributions

Researchers associated with the NextAIRE consortium have contributed to a new research preprint exploring how machine learning and environmental data integration can improve the prediction of atmospheric particle number size distributions.

The study, “Environmental and Spatiotemporal Data Fusion for Particle Number Size Distribution Prediction,” was made available on SSRN on 31 August 2026.

The research brings together M. J. Jiménez-Navarro, Ana Rodríguez López, Mario Lovrić, Simonas Kecorius and María del Mar Martínez Ballesteros.

Combining Environmental Data and Machine Learning

Understanding the number and size of airborne particles is essential for assessing air pollution and its potential impacts on human health and the environment.

The researchers developed a predictive framework combining hourly measurements from 36 monitoring stations worldwide with environmental, geographical and temporal information.

Different machine learning approaches were evaluated to predict particle number size distributions, including models based on detailed particle-size measurements and simplified mathematical representations.

Key Research Findings

The study highlights several important findings:

  • Human activities, land use and tourism-related factors influence atmospheric particle patterns.
  • Seasonal and daily variations play an important role in particle concentrations.
  • Combining different machine learning models can improve the robustness of predictions.
  • Different modelling approaches offer trade-offs between predictive accuracy and physical consistency.

Advancing AI-Driven Air Quality Research

The findings demonstrate the potential of combining environmental observations, geographical information and artificial intelligence to better understand atmospheric particle behaviour.

This research is closely aligned with NextAIRE’s objectives of advancing AI-based air quality research, environmental data analysis and interdisciplinary scientific collaboration.

The study is currently available as a research preprint on SSRN.

Read the Research Preprint

Title: Environmental and Spatiotemporal Data Fusion for Particle Number Size Distribution Prediction

Authors: M. J. Jiménez-Navarro, Ana Rodríguez López, Mario Lovrić, Simonas Kecorius and María del Mar Martínez Ballesteros

Platform: SSRN, 2026

Date posted: 31 August 2026

Full preprint: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=7366096