Skip to content

Multi-model benchmarking of PEcAn reanalysis against CMIP6 and TRENDY using ILAMB #4038

Description

@Tejas7007

Follow-on to #4019 (PEcAn-to-ILAMB conversion). That issue produced the conversion pipeline; this one covers benchmarking the converted PEcAn reanalysis against the CMIP6 and TRENDY model ensembles, and scoring the individual PEcAn SDA members.

Scope:

  • ILAMB configuration scoring PEcAn against observational benchmarks (Xu & Saatchi 2021 biomass, GIMMS LAI4g, HWSD2 soil carbon) over North America.
  • Build scripts for the CMIP6 ensemble (historical, and a historical+ssp245 splice) and the TRENDY (GCB 2024) ensemble, regridded to the PEcAn grid.
  • Per-member scoring of the 100 PEcAn SDA members to compare ensemble spread against the model ensembles.
  • Two evaluation windows (2012-2014 full CMIP6 sample; 2015-2023 longer record) reported together.

Out of scope (planned follow-ons): ensemble calibration / probabilistic scores (rank histograms, reliability); regional breakdowns; webpage hosting of the scorecard.

Deliverable: scripts, config, screening manifest, README, and a spread figure under modules/benchmark/inst/ilamb/.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions