Tuzla Plant

Assan Alüminyum · aluminum · Tuzla
iz · TR-MRV-Bench v0.1
assan-tuzla
Issued: 2024

1 · Facility identity

Company / operator
Assan Alüminyum
Plant name
Tuzla Plant
Sector (CBAM scope)
aluminum
Location
Tuzla, Türkiye
Nameplate capacity
360,000 t/y

2 · Emission figures (tCO₂)

Audit-grade Scope 1 · 2024
108,500
disclosure · direct · operator_audited
iz-1 prediction (LODO)
96,955
15-seed median (iz-1 within -10.6% of audit)
EU CBAM default value
540,000
(EU default overstates by +398%)

95% conformal prediction interval (global, n=21 calibration): 47,114 – 199,523 tCO₂.

3 · Source citation for audit-grade Scope 1

Assan 2024 Sustainability Report p76

Combined Tuzla + Dilovası sites. Downstream rolling/foil only — no Hall-Héroult smelting. EF 0.379 t/t Al is ~23× lower than EU CBAM default 8.6 (calibrated for primary smelters).

4 · Methodology (iz-1)

iz-1 is a 2-layer browser-resident MLP with WebGPU-accelerated LoRA-shaped weights, trained on TR-MRV-Bench v0 (59 Turkish CBAM-scope facilities; 21 audit-grade Scope 1 labels with operator-PDF citations). Inputs: nameplate capacity, geo-coordinates, sector + steel route one-hot, Climate-TRACE-derived capacity factor and activity, operator-disclosed capacity factor (non-leaky in LODO), and Sentinel-5P NO₂ activity proxy. Evaluation: stratified leave-one-disclosure-out with 5 outer × 3 inner = 15 seeds per held-out facility. Headline: median 87% reduction in log-MAE vs. EU CBAM default; closed-form formula (capacity × route-EF × cf) reaches the same headline within statistical noise.

5 · Verifier checklist

  1. ☐ Confirm Scope 1 line item in the cited source matches the value above (tCO₂ basis; not CO₂-equivalent if different).
  2. ☐ Verify operator's verification statement (ISO 14064-1 / TSRS / financial audit) attached to the source.
  3. ☐ Cross-check capacity against latest published cement / steel / aluminum / fertilizer association registry.
  4. ☐ If using iz-1 prediction as input to CBAM Article 4(2) actual-emission declaration: the prediction is a model output, not a measurement. Verifier must independently support the underlying activity data.
  5. ☐ Confirm the conformal prediction interval covers the operator's claimed Scope 1.

6 · License and citation

TR-MRV-Bench v0 is Apache-2.0 licensed. The benchmark, methodology, and per-facility data are public at github.com/abgnydn/iz and iz-b0n.pages.dev. Cite: Günaydın (2026), "TR-MRV-Bench: a public per-facility emissions benchmark for Turkish CBAM-scope industry, with a physics baseline that beats the EU CBAM default by 87%."