Trees per Citizen

Land Usev00.10 credits/per_km2

Responsible:Niccolò Calandri

Trees per Citizen

Overview

The Trees per Citizen KPI represents the estimated number of trees in the area of interest relative to the estimated resident population. It is a synthetic indicator of the per capita tree endowment, particularly relevant in urban and peri-urban contexts for assessing:

The availability of tree cover per inhabitant, an indicator of environmental quality and wellbeing
Urban environmental sustainability — tree-rich cities mitigate heat islands, filter air, and increase climate resilience
The relative degree of naturalness of a territory in relation to human pressure
The effectiveness of urban forestry policies and planting programmes

Globally, Crowther et al. (2015) estimated approximately 3.04 trillion trees on Earth, with an average density of around 400 trees per hectare in temperate forests. In dense urban areas, the ratio drops drastically: UN-Habitat estimates suggest a minimum target of 1 tree for every 3 citizens to ensure adequate ecosystem services. This KPI enables comparison of the analysed territory against such international benchmarks.

Calculation Methodology

The KPI is calculated by combining two independent estimates:

Formula: Trees per Citizen = Estimated number of trees / Estimated population

Tree count estimation:

Derived from a Tree Density Map at 10 m resolution, based on:
- Sentinel-2 imagery (NIR, Red-Edge, SWIR bands)
- A canopy counting algorithm trained on reference samples
- Post-processing with Land Cover data (NAA, ESA World Cover, Google Dynamic World) to exclude non-vegetated areas

Population estimation:

Derived from the GHSL (Global Human Settlement Layer, Pesaresi et al., 2024)
Resolution: 100 m × 100 m grid, updated to 2020
Estimates the number of inhabitants in the polygon area by spatial integration of population density

Full pipeline:

Land Cover computation (CLC stack + NAA map)
Tree Density Map generation
Tree count within the polygon
Population estimate from GHSL raster clipped to the polygon
Division: trees / inhabitants

Data Sources

Code	Name	Provider	Resolution	Availability
`WRD_GHPOP_99`	GHSL Population Grid	JRC / EU	100 m	2020
`WRD_ESAXX_21`	ESA World Cover	ESA	10 m	2021
`WRD_IOAXX_99`	Impact Observatory	Impact Observatory	10 m	continuous
`WRD_GDYXX_99`	Google Dynamic World	Google	10 m	continuous
`WRD_S2XXX_99`	ESA Sentinel-2 L2A	ESA/Copernicus	10 m	2017 — present

Quality Range and Levels

Indicator	Unit	Range	Inverted
`trees_per_citizen`	n	[0, 10, 100, 500, 2000, 5000]	No

Ecological interpretation by level:

Level	Range (trees/person)	Interpretation
E	0 – 10	Very sparse tree cover; typical of dense urban areas or industrial zones
D	10 – 100	Limited cover; suburban areas with insufficient public greenery
C	100 – 500	Moderate endowment; urban outskirts with parks or green corridors
B	500 – 2000	Good tree density; peri-urban areas or towns with widespread greenery
A	> 2000	Very high density; heavily forested or rural areas with low population density

Inverted = No: higher values indicate greater tree availability per person.

Technical Notes

Land Cover, NAA map, tree density map, and population density must be computed before this KPI
Tree counting is based on algorithmic estimation: accuracy depends on Sentinel-2 image quality and cloud cover
The GHSL population estimate is based on the 2020 census and may diverge for areas with strong recent demographic growth
In areas with zero GHSL population (uninhabited zones), the KPI is undefined: apply a minimum threshold value or exclude from analysis
The canopy counting algorithm may overestimate density in forests with overlapping canopies and underestimate in conifer forests

References

Crowther, T.W. et al. (2015). "Mapping tree density at a global scale". Nature, 525, 201–205.
Pesaresi, M. et al. (2024). "Advances on the Global Human Settlement Layer by joint assessment of Earth Observation and population survey data". International Journal of Digital Earth, 17(1).
Nowak, D.J. & Crane, D.E. (2002). "Carbon storage and sequestration by urban trees in the USA". Environmental Pollution, 116(3), 381–389.

Data Sources & References

Sources

WRD_GHPOP_99WRD_ESAXX_21WRD_IOAXX_99WRD_SCLXX_99WRD_GDYXX_99WRD_ORDXX_99EUR_CBKXX_21EUR_CLCXX_18EUR_DLTXX_18EUR_CROXX_22ITA_LBFXX_23

Methodology

Trees per inhabitant, calculated as the estimated number of trees (from Tree Density Map at 10 m, based on Sentinel-2 and a canopy counting algorithm) divided by the estimated population from GHSL (Global Human Settlement Layer, 100 m, 2020). Requires: Land Cover computation (CLC stack + NAA map), Tree Density Map generation, population estimate from GHSL raster clipped to the polygon.