Validate UHI_MODEL + THRESHOLD_MODE env vars; warn + fallback on typos

Simplify: reuse shared WBGT+percentile utils, batch DB writes, memoize reads

Surface per-zone thresholds on Vercel frontend

README: add UHI_MODEL × THRESHOLD_MODE env-var interplay table

Frontend: clarify threshold labels (city-wide default vs production per-zone)

UHILSTCorrector: defer to synthetic UHICorrector for LST-uncovered zones

Zone-specific trigger thresholds behind THRESHOLD_MODE env var

Dockerfile: COPY data/landsat_lst/ for LST UHI model features

Fix FEATURES_PATH resolution for Docker (COPY to /app) vs local

Add LST-anchored UHI corrector behind UHI_MODEL env var flag

Fix TOA accumulation window: 6h → 1h to match ERA5/GraphCast

pipeline: one-shot reconnect on DB stale connection

graphcast: handle static vars without a time dim (GFS path)

gfs: analytical TOA forcing + pipeline traceback logging

Fix GFS shortName regression: use ecCodes names, not cfgrib names

Add GFS init source as alternative to ERA5T (Phase 2 wiring)

Cut dead neural/LSTM pricing + predictor paths

Sync portfolio copy with post-session architecture

NASA POWER ingest: drop all-missing days instead of inserting empty rows

predictions ON CONFLICT: overwrite xgb/lstm/confidence/ensemble too

Cut XGBoost/LSTM heat predictor from pipeline and demo

Refresh Neon connection at the start of each pipeline run

heat_indices + healed_readings ON CONFLICT: overwrite all derived fields

Fix real Claude cost bug: token double-count, revert Haiku swap

Pipeline hardening: Haiku, dead code cleanup, consec fix, ERA5-lag init

Prewarm GraphCast at FastAPI startup to avoid cold-start fallback

NASA POWER ingest: heal NULL fields, request precip, respect 2-day lag

Fix stale Neon connection on Space wake

De-jargon pipeline cost panel: drop Claude model-name selector + token pricing

Simplify: drop narrative comments, clean arbitrary Tailwind values

Desktop hero polish + redesign ACTIVE banner

Mobile hero polish: single-KPI cards + document two-tier threshold

Mobile hero: timeline left, description + output stacked right

Mobile home page: 2-col pipeline + output; update REBUILD guide

Mobile polish: vertical timeline hero on Dashboard

Mobile responsive: hamburger sidebar + responsive grids

Frontend: use burn analysis values from DB, drop old neural pricing formulas

Fix pipeline failures: api.py import, NASA POWER resilience, CPU torch

Clean up merge conflict markers in neural_actuarial.py (reference code)

Merge branch 'main' of https://huggingface.co/spaces/jtlevine/climate-risk-engine

Replace neural actuarial model with GraphCast triggers + empirical burn analysis

Wire retrained trigger head into pricing (override alert/payout probs)

Add retrained trigger head (V1): 8% → 28% hit rate on insurance benchmark

Full revert to April 14 working state (config.py)

Full revert to April 14 working state (src/pipeline.py)

Wire classify_worker_advisory into risk_level output (no schema changes)

Revert src/database/schema.py to pre-benchmark state (keep heat_index.py + config.py)

Revert src/database/crud.py to pre-benchmark state (keep heat_index.py + config.py)

Revert src/pipeline.py to pre-benchmark state (keep heat_index.py + config.py)

Benchmark-driven: daily-relative worker advisory (config.py)