← All researchMF-PRISM-FIRE-2026-01 / interactive guide

THE IDEA, MADE VISIBLE

Was the fire changing—or the view?

Change sensor visibility independently of fire activity. See how an apparent escalation can disappear.

Research reportObservation-Aware Wildfire Escalation

Calculated illustration · change the inputs to inspect the mechanism.

01 / 04Guided chapter

Fix the underlying activity

The synthetic incident begins with activity 100 and observation opportunity 0.20. Its expected detection count is therefore 20.

Relative fire activity

100

New observation opportunity

0.6

Expected detections: 60
Raw count ratio: 3×
Activity ratio: 1×

Expected detections change from 20 to 60. The 3× raw ratio combines a 1× activity change with altered observation opportunity.

Expected detections = activity × observation opportunity

TRY THIS

Keep activity at 100 and move observation opportunity from 0.20 to 0.80. Raw detections quadruple, while adjusted activity does not change.

What this experiment represents. Synthetic mean-count model with perfectly known observation opportunity. No real incident data, uncertainty calibration, operational forecast or emergency warning is produced.

Source manuscript & release ↗Read the full explanation ↓

JavaScript is off. The initial figure above and the complete written explanation below remain available. Interactive controls require JavaScript.

1. Fix the underlying activity

The synthetic incident begins with activity 100 and observation opportunity 0.20. Its expected detection count is therefore 20.

2. Change the view

Cloud cover, viewing geometry and timing affect what a sensor can observe. Increasing opportunity can increase detections even when activity does not change.

3. Compare compatible evidence

In this deliberately simple model, dividing detections by known opportunity recovers activity. The paper studies bounds and evidence when that opportunity is not perfectly known.

4. Respect what was known when

A retrospective decision must use information available at issuance, not later revisions. This synthetic teaching scene is not a validated wildfire alert or growth forecast.

Wildfire Intelligence · Research Paper

Observation-Aware Wildfire Escalation

Sharp thermal-growth bounds and arrival-time-valid evidence

Research reportMF-PRISM-FIRE-2026-01 / v0.2

Distinguish a change in fire activity from a change in what the sensors could observe. The paper develops graph-constrained activity contrasts and arrival-time-valid evidence, then tests them under stated synthetic models.

Read the paper Release & files

Cover of Observation-Aware Wildfire Escalation — Current public editionv0.2 · 2026-09-12

Identifier: MF-PRISM-FIRE-2026-01
Series: Wildfire Intelligence
Edition: Version 0.2
Length: 19 pages
Reserved DOI: 10.5281/zenodo.22726215 (record reserved)

Plain-language guide

3 minute explanation

What this paper is really saying.

The paper asks whether an apparent surge in satellite fire detections reflects more fire or merely a better opportunity for the sensors to see it.

wildfirethermal detectionsobservation bias

Satellite detections are not direct, uniform measurements of fire. Clouds, viewing geometry, revisit timing, sensor coverage, and delayed data availability all change what can be observed.

The proposed method compares fire activity only after accounting for relative observation opportunity. It also enforces arrival-time validity: a forecast or escalation decision may use only evidence that was actually available at that moment, not later revisions.

How it works

The argument, without the notation.

The paper’s technical details matter, but the basic route can be understood in three moves.

Model observation opportunity

Represent when and where a sensor could have produced a detection, rather than treating every missing detection as evidence of no fire.

Compare compatible counts

Use graph-constrained contrasts and conditional paired-count reasoning to separate activity change from exposure change.

Respect information arrival

Timestamp first availability and prevent later observations or revisions from leaking backward into an earlier decision.

What it adds

The useful takeaways.

Raw detection growth is not automatically fire growth.
Observation opportunity becomes part of the statistical evidence.
The framework is designed to produce auditable escalation evidence under stated assumptions, not an official warning.

Scope & limits

What this does—and does not—establish.

Current status

Conditional mathematical theory and synthetic evidence; no real-fire forecast has been validated. Research only; not an official emergency warning.

This report develops a conditional method and evaluates it within the stated evidence. It does not establish production performance, operational safety, or calibrated real-world predictive skill beyond that evidence.

No operational wildfire forecast has been validated.
The method depends on adequate exposure calibration and componentwise identifiability.
Repeated or dependent evidence must not be treated as independent support.

Why it matters

Why anyone should care.

A model that confuses sensor opportunity with fire growth can issue false escalation signals. This work makes observability part of the evidence rather than an afterthought.

Key terms

The vocabulary, decoded.

These are the terms needed to understand the claim. The full paper uses them more precisely.

Thermal detection

A sensor observation indicating heat consistent with active fire, not a complete fire perimeter.

Exposure

The opportunity a sensor had to observe a location under the measurement process.

Arrival-time valid

Using only data that were available when the decision was issued.

Identifiability

The ability to distinguish separate causes—such as activity and observation—from the available data.

Independent review

Where scrutiny should concentrate.

The strongest review is not a general reaction. It tests the steps most capable of changing the conclusion.

01
Relative exposure calibration and componentwise identifiability.
02
Conditional paired-count assumptions and repeated-evidence handling.
03
Observation time versus first availability and later revisions.

Publication record

Go from explanation to evidence.

This page is a reading guide, not a substitute for the manuscript. The public record links the explanation to the paper, source package, review materials, and persistent identifier.

GitHub release Source folder

Document: Research Paper
Review status: Open for independent review
Published: 2026-09-12
DOI status: Reserved · 10.5281/zenodo.22726215
Canonical record: MF-PRISM-FIRE-2026-01

Metriq PRISM Laboratory, Observation-Aware Wildfire Escalation, Metriq PRISM Laboratory Research Paper MF-PRISM-FIRE-2026-01, Version 0.2, 2026. Corresponding contributor: Daniel H. Jeffery, ORCID 0009-0001-1200-6042. DOI: 10.5281/zenodo.22726215.