coral-reef-utac

GenesisAeon Package 114 — NOAA Coral Reef Watch's real, official Degree Heating Week (DHW) coral-bleaching threshold system, and the real, escalating series of four documented global bleaching events. Deliberately has no UTAC/CREP/AFET bridge and no invented Γ value — see DISCLAIMER.md for why.

For a plain-language explanation of the same topic (German, no jargon, written for general audiences), see WHITEPAPER.md.

Where this package came from

Chosen from a shortlist of 5 candidate gap-topics proposed via an external AI dialogue about GenesisAeon's environment/climate/species- protection coverage (UTACnextSteps.txt: Grok, then a separate Claude-Code-Web session). That session's specific technical claims — three invented Γ_tipping values (0.48/0.62/0.45), a "Pressure- inverted UTAC" formula presented as if extracted from the codebase, a "strict GPL-3.0 template" claim, and a git branch presented as if already created — did not survive independent verification (the formula appears nowhere in the actual code; the license claim was generalized from a single file, 64 packages are actually MIT vs. 7 GPL-3.0; the branch does not exist in any local repo). Both sessions converged on the topic being sound; this package is built from scratch with independently verified citations, not the other session's numbers.

Deliberately not one-sided

is_bleaching_response_uniform_across_genotypes() returns False — Walker et al. (2023) found real, measured variation in heat resistance between individual coral genotypes, with real fitness tradeoffs (resistant genotypes grow slower during recovery). A DHW threshold crossing does not mean uniform bleaching or uniform mortality.

What's real here

• NOAA Coral Reef Watch — Degree Heating Week (DHW) — the actual operational metric used to issue global bleaching alerts: 0–20°C-weeks accumulated over a 12-week window, 5 alert levels (4/8/12/16/20°C-weeks) each with a documented real-world consequence. The scale itself had to be expanded from 2 to 5 levels in December 2023 because 2023's heat stress exceeded what the original scale described.
• Hughes et al. (2017, Nature) — the 2015–2016 event, the third documented global bleaching event, hit 91% of individual Great Barrier Reef reefs.
• Hughes et al. (2018, Science) — a real 100-site global analysis (1980–2016): the median return interval between severe bleaching events at a given reef has fallen to just 6 years — too short for many coral communities to fully recover.
• NOAA (2024) — confirmed the fourth global bleaching event: 83.7% of the world's reef area affected, Jan 2023–Apr 2025. The full documented series is real and monotonically escalating: 21% (1998) → 37% (2010) → 68% (2014–2017) → 83.7% (2023–2025).
• Guillermic et al. (2021, Science Advances) — a real lab finding: corals can actively compensate for ocean acidification at normal temperature (28°C), but heat stress (31°C) disables that compensation entirely — not just an additive second stressor.
• Walker et al. (2023, Evolutionary Applications) — the honesty-check citation: real genotype-level variation in bleaching resistance in Acropora hyacinthus (Palau), with real fitness costs for resistant genotypes.

Quickstart

pip install coral-reef-utac

from coral_reef_utac import (
    alert_level_for_dhw, consequence_for_alert_level,
    GLOBAL_EVENTS, is_severity_monotonically_increasing, recovery_window_years,
    can_corals_compensate_for_acidification_under_heat_stress,
    is_bleaching_response_uniform_across_genotypes,
)

print(alert_level_for_dhw(9.5))                 # 2
print(consequence_for_alert_level(2))
print(list(GLOBAL_EVENTS.reef_area_affected_pct_by_event.values()))  # [21.0, 37.0, 68.0, 83.7]
print(is_severity_monotonically_increasing())    # True
print(recovery_window_years())                   # 6.0
print(can_corals_compensate_for_acidification_under_heat_stress())  # False
print(is_bleaching_response_uniform_across_genotypes())              # False

Development

pip install -e ".[dev]"
pre-commit install
ruff check src tests
mypy src
pytest

Citation

See CITATION.cff and .zenodo.json.