✕
Honesty ledger — assumptions & limits
• This is a lower bound on transport: A
single-layer, ground-level annual-average SR response under-represents long-range
transport, so every "share of burden far from source" figure here is conservative —
the real spread is wider, not narrower. For the same reason there is deliberately no
wind-rose or directional plot on this page: the matrix is an annual average and does
not resolve direction.
• Distance is to the NEAREST source cell, which is the honest cheap
metric and also its own limitation: a far cell is often downwind of many sources at once,
and gets filed under its closest one. The bands describe geometry, not attribution.
•
Present burden, not a repeal forecast: this page shows what the
currently-operating US coal plants' real 2025 emissions
contribute today. The separately-labelled §111-repeal delta scenario lives on the
national tile, not here.
• Nearest-centroid raster (MAUP):
• y0 (baseline mortality rate):
• All-ages population convention:
National 25+-consistent correction factor ≈ — which is why
the national figure is shown as a band.
• GEMM is an upper-bound CRF:
• Non-US remainder is off-map:
• GEMM risk threshold:
• Climate TRACE input-layer caveats:
• Why this page exists:
• VSL is illustrative: VSL = (2023 USD,
EPA-anchored central value) — prefer a range around the central figure, never "cost of
deaths" or "lives worth."
• Cell mapping:
• Source markers are grid cells, not individual plants: the neutral
markers show aggregated source grid cells. That is a property of
this page's data, not a withholding decision — the national payload behind it carries
per-cell emissions and a count of units per cell, and no per-unit rows at all, so grid
cell is the finest level this page can report. Per-plant detail for the Ohio Valley
cluster is on the source map, where the payload has it.
• No demographic or environmental-justice view — and why: there is
zero demographic data on this box, so nothing on this page speaks to who bears the
burden by race, income or vulnerability. Building that honestly would need (a) US
Census/ACS race-and-income tables, (b) EPA EJScreen indicators, (c) CDC/ATSDR Social
Vulnerability Index, and (d) a defensible spatial join from this variable-resolution
receptor grid down to tract or block-group level — an area- or population-weighted
(dasymetric) transfer, not a naive overlay. It would also need its own caveats: per-state
y0 cannot resolve within-state mortality differences, so a naive version would understate
real disparity, and attributing a group-level value to individuals is an ecological
fallacy. Not attempted here rather than approximated.