
How green is my city?
PLACE data can be used to map green spaces, places essential for bio-diversity and people’s well-being.

In many countries there is a lack of information on green spaces, where they are, how big they are and who they serve. Mapping the location, area and type of green space addresses SDG 11.7 which seeks to provide universal access to safe, inclusive and accessible, green and public spaces, for women and children, older persons and persons with disabilities.
So how would you go about answering the question "how green is my city?"
Well we start with PLACE data. Given its high resolution you easily see trees and green spaces. As PLACE imagery is true color, trees and grass are recognizable as such.
By way of example here's a tree detection exercise in Lilongwe, Malawi. This used a pre-trained model, meaning detection would only get better with more labels (i.e. someone works across the imagery tagging what they observe to be trees, this labeling process makes the model better at finding trees).

Tree count, Lilongwe. Green boxes are what the detection model determines to be a tree.
In Zomba, Malawi we used Esri's Text_SAM model to find trees. The video below shows the model in action:
Using the Text_SAM model to detect trees.
Within this area (a small section of Zomba was chosen for this demonstration) which measures approximately 0.3 sq km, SAM detected well over 1,000 trees with a estimated canopy cover of 0.05 sq km. This means this section of the city alone is 17% green and that's not including parks.
City planners in Zomba may be to able infer that the city has green lungs.