CREATORS: Jérémy Bernard (1,3), Erwan Bocher (2), Elisabeth Le Saux Wiederhold (3), François Leconte (4), Valéry Masson (5) and Camille Noûs (6)

AFFILIATIONS: 
(1) University of Gothenburg, Department of Earth Sciences, Sweden 
(2) CNRS, Lab-STICC, UMR 6285, Vannes, France
(3) Université Bretagne Sud, Lab-STICC, UMR 6285, Vannes, France
(4) Université de Lorraine, INRAE, LERMaB, F88000, Epinal, France
(5) Météo-France and CNRS, CNRM, UMR3589, Toulouse 31057, France
(6) Laboratoire Cogitamus, https://www.cogitamus.fr/

TITLE: Estimated height of the OpenStreetMap buildings of 24 French communes using the GeoClimate Software (version 0.0.1)

DATE: Data downloaded from OpenStreetMap between 1st of July and 30th of September 2021

MODEL USED FOR PRODUCTION: GeoClimate (version 0.0.1 - https://doi.org/10.5281/zenodo.5534680)

ABSTRACT: This dataset is related to a research manuscript proposed to GMD journal. It contains a folder called 'toReproduceResults' dedicated to any person willing to reproduce the main results presented in the manuscript. It also contains 24 folders, each of them corresponding to a French city, containing the building geometry footprints and their corresponding building height as well as averaged building height value aggregated at rectangular grid cell (100 m by 100 m). The footprint geometries comes from the OpenStreetMap project and the building height has been estimated using a RandomForest algorithm using as independent variables indicators describing the building size and shape and the building environment. The data has been produced using the GeoClimate Software (version 0.0.1).

CONTENT: 
If you are interested in reproducing the results, see the 'Readme.txt' which is located in the 'toReproduceResults' folder.

If you are interested by the datasets resulting from the research work, for each city, two geojson files called "BuildingScale" and "GridCellScale" are available in each city folder.
  - BuildingScale contains several fields:
	- ID_BUILD: Building unique identifier
	- THE_GEOM: Footprint geometry of the building (comes from OpenStreetMap data)
	- HEIGHT_ROOF: Building's roof height estimated by the RandomForest model (if the field "ESTIMATED" = 0, then the data is not estimated but comes from OpenStreetMap)
	- HEIGHT_WALL: Building's wall height estimated by the RandomForest model (if the field "ESTIMATED" = 0, then the data is not estimated but comes from OpenStreetMap)
	- ESTIMATED: Whether or not the building height has been estimated (ESTIMATED = 1) or comes from the OpenStreetMap data (ESTIMATED = 0)

  - GridCellScale contains several fields:
	- ID_RSU: Grid cell unique identifier
	- THE_GEOM: Grid cell geometry (100 m x 100 m square)
	- AVG_HEIGHT_ROOF: Averaged building's roof height within the grid cell (a building intersecting several grid cell being attributed to the one it shares the biggest area)
	- STD_HEIGHT_ROOF: Standard deviation building's roof height within the grid cell (a building intersecting several grid cell being attributed to the one it shares the biggest area)
	- AVG_ESTIMATED: Average building "ESTIMATED" value (the closer from 1 the more buildings in the grid cell have been estimated from the Random Forest model)
	- EFFECTIVE_TERRAIN_ROUGHNESS_LENGTH: Effective terrain roughness length (z0) as defined by Hanna and Britter (Hanna, Steven R., and Rex E. Britter. Wind flow and vapor cloud dispersion at industrial and urban sites. Vol. 7. John Wiley & Sons, 2010.). For more information, refer to GeoClimate v. 0.0.1 documentation (https://zenodo.org/record/5534680/files/GeoClimate_0.0.1_documentation.pdf?download=1)
	- GROUND_SKY_VIEW_FACTOR: Ground Sky View Factor such as defined by Stewart et Oke (2012): ratio of the amount of sky hemisphere visible from ground level to that of an unobstructed hemisphere. In our case, only buildings are considered as obstructing the atmosphere. For more information, refer to GeoClimate v. 0.0.1 documentation (https://zenodo.org/record/5534680/files/GeoClimate_0.0.1_documentation.pdf?download=1)