Comparison between local climate zones maps derived from administrative datasets and satellite observations

Abstract : This paper proposes a method for generating maps of Local Climate Zones (LCZs) within a GIS using administrative and 2.5D building databases. The LCZs are computed from morphological indicators and building typology, on vector reference spatial units that correspond to urban islets, i.e. blocks of buildings surrounded by nearby roads. The main originality is that, while mean building height criteria correspond exactly to the LCZ classification, a k-means statistical method is used to determine, for each city, the limits between compact and open (and sparsely built) LCZs for high-, mid- and low-rise LCZs, respectively. For example, in SO12 LCZ look-up tables and the WUDAPT-L0 mapping approach, “compact” LCZs correspond to a building density of over 40%. The resulting groups for Nantes, Toulouse and Paris for mid-height, treated with the proposed statistical method, are 36%, 37% and 33.8% respectively. The LCZ maps for these three cities are compared to the WUDAPT LCZ maps, the latter being obtained from satellite imagery at a re- solution of 100 m. MApUCE LCZ maps show more spatial details, due to their finer resolution, and more variety in urban LCZs within each conurbation. This is very important for modeling micro-climatic effects on town peripheries.
Type de document :
Article dans une revue
Urban Climate, Elsevier, 2019, 27, pp.64-89. 〈10.1016/j.uclim.2018.10.004〉
Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-01930505
Contributeur : Gwendall Petit <>
Soumis le : vendredi 7 décembre 2018 - 15:53:29
Dernière modification le : vendredi 7 décembre 2018 - 16:27:17

Fichier

LCZ-MAPUCE-WUDAPT_20180727.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

Citation

Julia Hidalgo, Guillaume Dumas, Valéry Masson, Gwendall Petit, Benjamin Bechtel, et al.. Comparison between local climate zones maps derived from administrative datasets and satellite observations. Urban Climate, Elsevier, 2019, 27, pp.64-89. 〈10.1016/j.uclim.2018.10.004〉. 〈hal-01930505〉

Partager

Métriques

Consultations de la notice

103

Téléchargements de fichiers

48