Forest roads (Congo Basin)

Description

Road development in the Congo Basin forest is continuously monitored from 2019 onwards in high spatial and temporal detail. A deep learning method is applied to 10 m scale Sentinel-1 and Sentinel-2 imagery for automated road detections on a monthly basis. This version presents 6 years of road development (57,363 km) from 2019-2024. 

The data is composed of line features distributed in .shp and .geojson formats. The following attributes are stored for the line features:

NetworkID: A unique ID for each connected road network.

SegLenM: The length of the road segment (in meters).

NetLenM: The length of the connected road network (in meters).

Month: The road segment opening month.

Year: The road segment opening year.

MonthNum: The road segment opening month, depicted as a continuing count since the start of monitoring (e.g. 13 = January 2020). This attribute can be used for smooth and continuous temporal analyses or visualizations.

Additional information

More information about the forest road mapping project can be found at: https://wur.eu/forest-roads

Continuously updated road maps can be interactively viewed  at: https://nrtwur.users.earthengine.app/view/forest-roads

The dataset can be accessed in Google Earth Engine at: ee.FeatureCollection('projects/wurnrt-loggingroads/assets/distribution/forestroads_afr_2019-01_2024-12')

The scientific paper (Slagter et al., 2024) describing the methods to produce this dataset can be found at: https://doi.org/10.1016/j.rse.2024.114380

Citation

Please cite the following when referring to this dataset:

Slagter B., Fesenmyer K., Hethcoat M., Belair E., Ellis P., Kleinschroth F., Peña-Claros M., Herold M., Reiche J. (2024). Monitoring road development in Congo Basin forests with multi-sensor satellite imagery and deep learning. Remote Sensing of Environment.