Skip to content

JSON Schemas

The pipeline emits several JSON files containing structured metadata and QA metrics. These files allow downstream tools to audit processing decisions and validate outputs.

This page summarizes the purpose and structure of each schema.

1. ENVI export metadata

Recorded in:

*_export_metadata.json

Contains:

  • wavelength array
  • band names
  • reflectance scaling (NEON conventions)
  • mask types and bit fields
  • CRS and affine transform

2. Topographic correction metadata

*_topo_metadata.json

Includes:

  • slope and aspect statistics
  • solar geometry used
  • mask adjustments
  • correction parameters

3. BRDF metadata

*_brdf_metadata.json

Includes:

  • BRDF coefficients per wavelength
  • RMSE of BRDF fits
  • flags for unstable fits
  • view/sun geometry summaries

The streamlined BRDF coefficient payload is also written as:

*_brdf_model.json

This model JSON contains:

  • iso, vol, and geo coefficient tables
  • kernel names used during fitting/application
  • geometric kernel parameters such as b_r, h_b, and solar_zn_type
  • ndvi_binning_enabled, indicating whether NDVI-stratified BRDF bins were used during fitting
  • ndvi_edges, which stores the realized NDVI bin boundaries (n_bins + 1 values) used to stratify the coefficient rows

ndvi_edges is metadata for interpreting the BRDF model. It explains which NDVI stratum each coefficient row belongs to; it is not a per-pixel NDVI product. When NDVI binning is disabled, the model stores a single neutral bin with ndvi_edges = [-1, 1].

4. Convolution metadata (sensor harmonization)

*_convolution_metadata.json

Contains:

  • SRF files used
  • wavelength alignment checks
  • brightness correction coefficients
  • bandpass integration statistics

5. QA metrics schema

The QA JSON file contains:

  • reflectance summary statistics
  • mask percentages
  • wavelength metadata
  • BRDF/brightness coefficients
  • geometry metadata

Exact fields are defined in the validation section.

Using schemas in downstream workflows

These JSON files support:

  • reproducibility
  • debugging
  • statistical evaluation of harmonization quality
  • provenance tracking for scientific analyses

Next steps