Researcher

Nuno Carvalhais

Nuno Carvalhais develops model-data approaches for Earth-system research. His joint work includes using deep learning to forecast wildfire danger from environmental observations. Carvalhais has led Max Planck’s Model-Data Integration Group since 2012. His group works on ecosystem modeling, data assimilation and hybrid models. Carvalhais coauthored Deep Learning Methods for Daily Wildfire Danger Forecasting with seven other researchers. The original paper compares learned spatial and temporal models with a random-forest baseline using Earth-observation data. Carvalhais leads a group in biogeochemical integration. His group focuses on terrestrial ecosystem dynamics, model-data fusion, vegetation remote sensing and hybrid modeling.

Photo of Nuno Carvalhais

Timeline

  1. 2012

    Model-data integration group

    Carvalhais has led Max Planck’s Model-Data Integration Group since 2012. His group works on ecosystem modeling, data assimilation and hybrid models.

  2. 2021

    Learning wildfire danger

    Carvalhais coauthored Deep Learning Methods for Daily Wildfire Danger Forecasting with seven other researchers. The original paper compares learned spatial and temporal models with a random-forest baseline using Earth-observation data.

  3. As of October 2026

    Earth-system research

    Carvalhais leads a group in biogeochemical integration. His research focuses on terrestrial ecosystem dynamics, model-data fusion, vegetation remote sensing and hybrid modeling.

Sources and credits

Photo: www.ellis-jena.ai; Nuno Carvalhais profile photograph; creator not stated. License: Not stated by source. Source: ellis-jena.ai. Cropped to a square and resized. Ask us to remove or replace this photo.

  1. Nuno Carvalhais – ELLIS Unit JENA ellis-jena.ai. Checked 13 October 2026.
  2. [2111.02736] Deep Learning Methods for Daily Wildfire Danger Forecasting arxiv.org. Checked 13 October 2026.
  3. Nuno Carvalhais, Dr. | Max Planck Institute for Biogeochemistry bgc-jena.mpg.de. Checked 13 October 2026.