ARGUS Open methods
Methodology

How the numbers are made.

For each analytical capability: what the model actually computes, which data feed it, how it has been validated, and — deliberately — its known limits.

v1 · July 2026This page is versioned and updates as the models change.
Traceability. Every figure on this page is traceable to the production codebase or a named validation run. Where no defensible figure exists, none is shown — a limit stated plainly beats a number invented confidently.
  1. 01 Shock Severity Index
  2. 02 Earthquake impact engine
  3. 03 Per-building damage
  4. 04 Food-security nowcast
  5. 05 Forecasts & the ledger
  6. 06 Convergence watch
  7. 07 Data boundaries & attribution
  8. 08 The honesty contract
01 · Shock severity

Shock Severity Index — detect, then quantify.

The SSI is a two-stage pipeline. Stage 1 monitors every watched country for the presence of an acute shock and scores it on six weighted factors — scope, criticality, duration, elasticity, interconnectedness, and policy-response capacity. Scope and duration are derived live from the event itself (hazard alert level, conflict status); the remaining four come from per-shock-type analyst priors. Stage 2 projects the economic impact of a detected shock: per-sector severities drive GDP, inflation, fiscal, trade, currency and employment components through sector elasticities and per-country baselines, with cross-sector interaction terms.

Data sources

GDACS disaster alerts, a verified armed-conflict catalogue, WHO disease-outbreak notifications, and per-country macroeconomic baselines.

Validation

Detection is wired end-to-end for two shock types today — natural disasters and geopolitical conflict. Stage-2 projections are being calibrated against documented historical shocks; until that calibration is signed, Stage-2 outputs are presented as projections, not measurements.

Known limits
  • Several factor priors are marked DRAFT in the configuration and are pending analyst sign-off. The SSI's constants are under active review; expect revisions.
  • Seven further shock types (beyond natural disaster and conflict) are specified but not yet wired to live detectors.
  • Stage 2 currently runs on deterministic elasticity means; formal uncertainty bands are a staged increment, not yet shipped.
  • The framing is economic and supply-chain by design — the SSI is not a humanitarian impact estimate.
02 · Seismic

Earthquake impact engine — real footprints first.

For a real earthquake with a published USGS ShakeMap, the engine intersects the actual intensity contours with administrative districts (geoBoundaries ADM2) server-side and reports per-district shaking plus population exposure by intensity band, using GHSL gridded population. For the window before a ShakeMap exists — and for hypothetical scenarios — it synthesises a footprint from an intensity-prediction equation, Allen, Wald & Worden (2012), extended with a finite-fault effective-distance correction built from Wells & Coppersmith (1994) rupture scaling and Thompson & Worden (2018) unknown-strike averaging.

Data sources

USGS ShakeMap intensity contours and PAGER; GDACS event footprints; geoBoundaries ADM2 districts; GHSL population.

Validation

The finite-fault correction was fitted against the published USGS ShakeMap for the M7.1 Venezuela earthquake: mean contour error fell from 21.6 km to 8.8 km, and the corrected model recovers the innermost MMI 8 ring that a point-source form misses. District exposure bucketing was checked against USGS PAGER for the M7.5 Venezuela event: the engine's MMI 6+ exposed population matched PAGER's published figure (8.0M vs 8.0M).

Casualty ranges

Fatality brackets use the Jaiswal-Wald (2010) lognormal lethality form with a global vulnerability bracket — the span of the range is the uncertainty. They are labeled modeled, order-of-magnitude: a planning signal, never a prediction. When USGS publishes PAGER for an event, PAGER is authoritative — the internal bracket is shown alongside it, never blended with it, never as a competing headline.

Known limits
  • The casualty model is not calibrated per country — it uses a global vulnerability bracket, which is why it reports a wide range rather than a point estimate.
  • Scenario footprints are idealised concentric contours from the intensity-prediction equation; a real published ShakeMap supersedes them the moment it exists.
  • Exposure counts derive from modeled gridded population, not census rolls; they are exposure to shaking, not affected-person counts.
  • The engine emits only what it actually computed — surfaces without a defensible value stay empty rather than showing a synthesised one.
03 · Building scale

Per-building damage — observed beats modeled.

At close zoom the globe estimates damage probability per building: Google Open Buildings footprints, the real ShakeMap ground-motion value sampled under each footprint, and a two-part model — a lognormal fragility curve that sets the absolute damage rate of a zone, plus a gradient-boosted ranking that orders buildings within the viewport by observable covariates (footprint size, local density, micro-terrain). Where satellite damage assessments exist, each assessed building's status is displayed as a fact, not a probability — and inside those assessed areas, observed data always wins; the model fills only the gaps between them.

Data sources

Google Open Buildings footprints; USGS ShakeMap ground-motion grids; commercial very-high-resolution damage assessments (Maxar, BlackSky, Vantor); Copernicus Emergency Management Service rapid-mapping gradings; radar-measured ground deformation (InSAR line-of-sight displacement) as a regional observed layer.

Validation

The fragility curve was re-fitted by maximum likelihood on three satellite-observed damage tiles — 46,564 buildings spanning peak ground accelerations of 0.25–0.84 g. The ranking model's honest skill: AUC 0.65–0.80 within an imaged zone, and ≈0.67 when transferred to an unimaged area — above the bare fragility curve's ≈0.58. An earlier coastal-amplification heuristic over-predicted observed damage roughly threefold and was removed: where the observed data contradicted the model, the observed data won.

Known limits
  • Outside an observed assessment area, every per-building value is a probability, labeled indicative — never a statement about an individual building's condition.
  • Calibration rests on a small number of tiles from one event region; transfer to different building stocks is the ≈0.67 new-area figure, not the higher within-zone one.
  • Satellite assessments are themselves model outputs of their providers, with their own error; Copernicus gradings cover only affected buildings, so their percentages read "confirmed among assessed", not a town-wide rate.
04 · Food security

Food-security nowcast — ninety days ahead.

A per-country nowcast of the 90-day trajectory of a food-insecurity proxy built from near-real-time phone-survey indicators (food-consumption and coping-strategy prevalence, normalised 0–100). The model is a three-part gradient-boosted ensemble exported to ONNX — a baseline regressor, a quantile model for uncertainty bounds, and a non-linear regime model — over 26 features computed from each country's own survey history: lags, rolling means and volatilities, trend, seasonality, and a lean-season flag.

Data sources

HungerMap LIVE survey history (the only feature source — no media or macro features), plus crop-calendar lean-season timing.

Validation

On its held-out test set the model reports MAE 1.6 percentage points, direction accuracy 97.7%, and R² 0.98. Read those honestly: the model is autoregressive — it extrapolates a country's own survey history — so these scores measure trajectory extrapolation, not shock anticipation.

Known limits
  • Inference is skipped for countries with under 90 days of stored survey history — the model never runs on synthetic inputs.
  • Coverage is bounded by where phone-survey collection operates; no surveys, no nowcast.
  • Being history-only, the nowcast is blind to exogenous shocks that have not yet moved the survey series.
  • In scoring, the nowcast can only raise a country's food-security signal, never suppress a live one.
05 · Accountability

Structural forecasts and the forecast ledger.

The structural electoral forecaster is a deliberately simple, fully inspectable logistic model over 11 structural features. It scores the probability that a governing party loses its next election, trained on 1,407 competitive national elections across 128 countries since 1990 (competitive defined as V-Dem electoral-democracy ≥ 0.5). It is a calibrated risk rank, not a crystal ball — its mean predicted probability equals the historical base rate (0.544) by construction.

Validation

Strictly out-of-sample only: forward-chaining AUC 0.714 (train on the past, predict the next) and leave-one-country-out AUC 0.721 (train on 127 countries, predict the held-out one). No random cross-validation — it leaks.

The ledger

Every call runs the same loop: predict → lock → resolve → score. A forecast is locked as an immutable snapshot the moment it is made — the number Argus is later graded on can never be retroactively edited. When the outcome is established, the entry is resolved and Brier-scored, misses included. Forecasts built from structural priors alone are flagged as such, and every entry carries its framing — "calibrated relative risk · watch flag" — rendered verbatim, never as "prediction". The track record is public, no sign-in required: /api/ledger/track-record.

Known limits
  • AUC ≈ 0.71 is honest skill for structural signals — materially better than chance, far from certainty. Treat outputs as ranked watch flags.
  • Structural features move slowly; the model does not see fast idiosyncratic events. Narrative signals are layered separately and flagged as such.
  • The ledger is young. The resolved-forecast count and the scores live on the public feed itself — this page states none, so it can never overstate them.
06 · Convergence

Convergence watch — a watch, not a prediction.

For each monitored country, six deterministic danger vectors — conflict, economy, currency, food, governance, regime — are computed from the live signals Argus already scores, plus V-Dem regime type. The convergence count is simply how many vectors are elevated at once. There is no model in this loop: the stacking is arithmetic, so it can be audited by counting.

Grounding

In a backtest over 4,841 country-years (1996–2024), countries with three or more simultaneously deteriorating vectors were 7.3× more likely to experience an irregular government change the following year than countries with one or none. Risk stacks monotonically with the count.

Known limits
  • The lift comes from additively stacking independent signals — the backtest found no non-linear "convergence magic", and we do not claim one.
  • Irregular government change is a rare event: even at high convergence, most country-years see no such change. The flag marks a danger window, nothing more.
  • Each vector inherits the limits of its underlying signal — including the media-derived nature of the conflict signal (see below).
07 · Data

Data boundaries and attribution.

Argus computes on top of public and licensed datasets, and the boundary of those datasets is the boundary of the product. Outside the monitored-country list, Argus has no live tools — and says so rather than improvising.

Administrative boundaries
geoBoundaries (ADM2), William & Mary geoLab — CC-BY. Used exclusively for district geometry.
Population
GHSL — Global Human Settlement Layer, European Commission JRC.
Building footprints
Open Buildings, Google Research.
Hazards
USGS (ShakeMap, PAGER, secondary hazards), GDACS, NOAA (tropical cyclones), Copernicus EMS (rapid mapping, InSAR ground deformation).
Food security
IPC analyses; HungerMap LIVE survey indicators; World Bank indicators.
Governance & regime
V-Dem, Freedom House, and curated analyst catalogues reviewed against public watchlists.
Economy & climate
Frankfurter exchange rates, Open-Meteo precipitation, satellite vegetation indices.
Media signal
GDELT global media monitoring — article volumes and deviation from baseline.

On conflict: Argus's conflict signal measures shifts in global media coverage. It is media-derived monitoring — labeled as such — and is not front-line reporting. No claim of ground truth is made where none exists.

08 · The contract

The honesty contract.

Four rules bind every surface of the product. They are enforced in the code paths that produce the numbers, not just promised here.

Modeled is never measured

Every modeled output is stamped as modeled, and order-of-magnitude estimates are labeled as such wherever they appear. A scenario can never be mistaken for an observation.

Observed wins inside its area

Wherever a satellite-observed assessment exists, it overrides the model inside its assessed area. Models fill gaps; they do not overrule evidence.

Authoritative sources lead

When an authoritative estimate exists — USGS PAGER for earthquake fatalities — it leads. Internal brackets are presented beside it as brackets, never blended into it.

Numbers are real or absent

When a feed is down or a value cannot be computed, the surface shows an honest empty state. Nothing is padded, backfilled, or invented to look complete.

Questions about a specific figure, or a limit we have not stated? admin@notamy.app — challenges to the methodology are welcome; they make it better.