Table 1.

Characteristics of the greenspace data sources

	NDVI*	OSM*	UA*
Type of measurements	Biophysical variables in vegetation	Data of categorical land uses	Data of categorical land uses
Type of data	Primary data	Primary data	Secondary data
Key characteristics	Using Landsat image to calculate the visible and near-infrared light reflected by vegetation. The result refers to the NDVI	A project which allows the community of Internet users to continuously update open and publicly available resources	Land cover data based on satellite imagery
Data availability	Worldwide provided	Worldwide provided, although with different degrees of accuracy	Data available for major urban cities in the European area
Time scale	Depends on the Landsat satellites	Continuously update	Data prepared periodically, currently from 2006, 2012 and 2018
Responsible agency	Landsat images available from the United States Geological Survey (USGS)	Database maintained by the OpenStreetMap Foundation	Project coordinated by the European Environment Agency (EEA)
Greenspace categories identified	Greenspace were identified with the use of supervised classification based on the representative samples for the different green space types in the digital image	– Allotments – Cemetery – Farmland/ farmyard – Forest/wood – Garden – Grassland – Greenfield – Greenhouse horticulture – Meadow – Nature reserve – Orchard – Park – Plant nursery – Scrub – Trees – Village green – Wetland	– Green urban areas (Code 14,100) – Arable land (annual crops) (Code 21,000) – Permanent crops (Code 22,000) – Pastures (Code 23,000) -Complex and mixed cultivation (Code 24,000) – Forests (Code 30,000) – Herbaceous vegetation associations (Code 32,000) – Wetland (Code 40,000)

^*NDVI: Measurement of calculating the Normalized Difference Vegetation Index (NDVI) by analysising Landsat satellite images; OSM: the OpenStreetMap dataset; UA: the European Urban Atlas dataset