GLobAl building MOrphology dataset for URban climate modelling

GLobAl building MOrphology dataset for URban climate modelling (GLAMOUR) offers the building footprint and height files at the resolution of 100 m in global urban centers.

the `BH_100m` contains the building height files where each file is named as `BH_{lon_start}_{lon_end}_{lat_start}_{lat_end}.tif`.

the `BF_100m` contains the building footprint files where each file is named as `BF_{lon_start}_{lon_end}_{lat_start}_{lat_end}.tif`.

Here `lon_start`, `lon_end`, `lat_start`, `lat_end` denote the starting and ending positions of the longitude and latitude of target mapping areas.

To avoid possible confusion, it should be clarified that the 'building footprint' in GLAMOUR represents the 'building surface fraction', i.e., the ratio of building plan area to total plan area.

We also offer the snapshot of source code used for the generation of the GLAMOUR dataset including:

`GC_ROI_def.py` defines regions of interest (ROI) used in the mapping of the GLAMOUR dataset.

`GC_user_download.py` retrieves satellite images including Sentinel-1/2, NASADEM and Copernicus DEM from Google Earth Engine and exports them into Google Cloud Storage.

`GC_master_pred.py` downloads exported data records from Google Cloud Storage and then performs the estimation of building footprint and height using Tensorflow-based models.

`GC_postprocess.py` performs postprocessing on initial estimations by pixel masking with the World Settlement Footprint layer for 2019 (WSF2019).

`GC_postprocess_agg.py` aggregates masked patches into larger tiles contained in the GLAMOUR dataset.