Downscaling urban thermal signals with machine learning: A case study of Danish functional urban areas

Maria Castro; João Paixão; Inês Girão; Bruno Marques; Rune M. K. Zeitzen; Rita Cunha; Caio Fonteles; Peter Thejll; Hjalte Jomo Danielsen Sørup; Quentin Paletta; Ana Patrícia Oliveira

doi:10.1016/j.uclim.2026.102947

May 25, 2026Urban ClimateOpen Access

Downscaling urban thermal signals with machine learning: A case study of Danish functional urban areas

View Full Paper

Ask AI

Bookmark

Share

Key Points

This research aims to enhance air temperature mapping accuracy in urban areas using machine learning techniques.
Developed a machine-learning downscaling model (UTS-D) for air temperature resolution.
Employed crowdsourced data and geospatial predictors to analyze neighborhood-scale temperatures.
Utilized Random Forest algorithm to maximize accuracy, achieving a mean absolute error (MAE) of 0.76 K.
Downscaled temperatures reveal distinct urban thermal signatures associated with land cover types and local climate zones (LCZ).
The UTS-D model outperformed the DANRA reanalysis in urban environments, but showed temperature overestimation in rural areas.

Ask about this paper

Answers are grounded in this paper and cite their sources.

Implication

Examines the effectiveness of a machine learning model for air temperature mapping in urban areas, indicating improved resolution.

Abstract

• Crowdsourced stations and geospatial predictors enable neighbourhood-scale air-temperature mapping. • A machine-learning downscaling model (UTS-D) for increasing air-temperature resolution. • Random Forest achieves highest accuracy (MAE of 0.76 K), with best skill during heatwaves. • Downscaled fields reveal clear urban thermal signatures linked to LCZ and land cover. • The model outperforms DANRA reanalysis in urban areas but overestimates rural temperatures.

Discussion

No takes yet. Share an insight, caveat, or question.

Cite This Study

Castro et al. (2026) studied this question.

synapsesocial.com/papers/6a17da9b3fad632b0f9d7a13 https://doi.org/10.1016/j.uclim.2026.102947

View Full Paper

Ask AI

Bookmark

Share

Downscaling urban thermal signals with machine learning: A case study of Danish functional urban areas

Key Points

Ask about this paper

Implication

Abstract

Discussion

Cite This Study

Downscaling urban thermal signals with machine learning: A case study of Danish functional urban areas

Key Points

Ask about this paper

Implication

Abstract

Discussion

Cite This Study

Keep reading

Keep reading