# Data Sources — Petrel Aridity Index Layer v0.1

All input data used to build the v0.1 vintage, with citations and licensing
notes. Every input is commercial-clean (Copernicus open data / public domain /
free-reuse-with-attribution) — no share-alike sources are included. v0.1 is a
**deterministic climatology**, not a trained model: there is no labelled
training set, only the ERA5 record and the two published method references
below.

## Climate inputs

### 1. ERA5 monthly 2 m temperature, 1940–2024

Global reanalysis 2 m air temperature, monthly means, 0.25°.

- **Citation:** Hersbach, H., Bell, B., Berrisford, P., et al. (2020). The ERA5
  global reanalysis. *Quarterly Journal of the Royal Meteorological Society*,
  146(730), 1999–2049. https://doi.org/10.1002/qj.3803. Data: Copernicus
  Climate Change Service (C3S) Climate Data Store.
- **License:** Copernicus / ECMWF open data — free reuse including commercial,
  with attribution to Copernicus C3S.
- **Used as:** the sole driver of Thornthwaite 1948 potential evapotranspiration
  (PET), over the full 85-year record.

### 2. ERA5 monthly total precipitation, 1940–2024

Global reanalysis total precipitation, monthly means, 0.25°.

- **Citation:** Hersbach et al. (2020), as above; Copernicus C3S CDS.
- **License:** Copernicus / ECMWF open data (free reuse incl. commercial, with
  attribution).
- **Used as:** the numerator P of the long-term aridity index P/PET.

This is the **same ERA5 record** the Petrel drought v2 pipeline consumes; the
aridity Layer is a byproduct of that ingest, at zero additional data cost.

## Method references (no data, published algorithms)

### 3. Thornthwaite 1948 PET

- **Citation:** Thornthwaite, C. W. (1948). An approach toward a rational
  classification of climate. *Geographical Review*, 38(1), 55–94.
  https://doi.org/10.2307/210739
- **Used as:** the temperature-only PET estimator (see `methodology.md` for why
  it is chosen over Penman-Monteith for the full-record backfill).

### 4. UNEP aridity classification (Middleton & Thomas 1997)

- **Citation:** Middleton, N., & Thomas, D. (1997). *World Atlas of
  Desertification* (2nd ed.). UNEP / Arnold, London.
- **Used as:** the five P/PET class thresholds (hyper-arid → humid) applied to
  the continuous index.

## Geographic mask (post-processing)

### 5. GSHHG v2.3.7 full-resolution shoreline (`gshhg_f`)

Petrel's canonical land mask — the Wessel & Smith Global Self-consistent,
Hierarchical, High-resolution Geography database at full resolution
(~100 m native).

- **Citation:** Wessel, P., & Smith, W. H. F. (1996). A global,
  self-consistent, hierarchical, high-resolution shoreline database.
  *Journal of Geophysical Research*, 101(B4), 8741–8743.
- **License:** LGPL (internal masking use only; not redistributed as data).
- **Used as:** the coast clip applied to all three rasters (`all_touched` — any
  cell touching real land keeps its value). GSHHG L1 only, so Antarctica is
  excluded by standard.

## Reproducibility

The exact source paths, method knobs, and promote configuration for v0.1 are
versioned in `configs/aridity_layer.yaml` in the Petrel ML repository. That
config is the source of truth — if anything here disagrees with it, the config
wins.