Going Inside the Pixel
In the world of forest monitoring, the pixel is king. Pixels depicting tree cover loss show us where forests may be in trouble from pressures such as logging, agriculture, and mining. By counting the pixels, we can estimate the extent of the damage. The pixels can show us areas damaged by fire, help us find new logging roads, and see where agricultural activity is degrading forests. Great, so is the problem solved? What do we do next?
Pixels are squares around 30 meters tall by 30 meters wide detected by satellites like Landsat or Sentinel. Analysis of this data can identify changes in the forest canopy. However it has limitations:
It doesn’t tell the story of what is actually happening on the ground.
The data is known to have errors and limitations, for example is is difficult to distinguish clearance of existing palm oil trees from natural forest clearance.
Most of the activity in the map above is replanting of palm trees. To go inside the pixel, we are using OpenStreetMap. In OSM volunteers trace upon donated satellite data to draw roads, villages, buildings, dams, palm tree orchards, and much more. Above all else, OSM is a community of volunteers interested in create free open maps of the world. It is used by groups such as the Red Cross to support humanitarian aid. We want to build this community for environmental mapping.
Brabanta, Democratic Republic of the Congo - mapped by Map for Environment in OpenStreetMap
With detailed mapping we can see exactly what is happening. In the example above we can see how palm plantations are swelling the population of local towns and villages that we theorize is indirectly straining nearby primary forests that are used for fuel and food. Here are some of things this data can tell us:
We can estimate the size of the palm plantation.
We can estimate the size of the population.
We can see the size of the palm mill’s storage tanks.
In Map for Environment we can overlay this data with the tree cover loss pixels and with government concession and title data.
This type of mapping needs a community, and needs support. The Bing and Mapbox Satellite imagery donated to OSM is often old and does not have complete coverage. We are looking to fill those gaps by purchasing new DigitalGlobe imagery.
How can you help?