CarbonMap logoCarbonMapWest Java biomass residue intelligence

Methodology

Reproducible Workflow

01

Multi-source data acquisition

Active workflow stage

Google Earth Engine exports Sentinel-2, VIIRS, ESA WorldCover, and cropland mask layers. These layers are paired with West Java boundaries, OpenStreetMap road features, agro-industrial facilities, and commodity production records.

Inputs

Sentinel-2VIIRSESA WorldCoverCropland mask

Outputs

Raster predictorsStudy boundaryFacility and road vectors

Study Snapshot

Grid

1 km

analysis resolution

Districts

27

West Java coverage

Cells

37,839

grid observations

Facilities

227

agro-industrial points

Interpretation Boundary

Suitability classes and deep-learning outputs are derived from proxy labels generated from a rule-based spatial suitability index. They should be interpreted as decision-support outputs, not field-validated measurements of residue availability.