Pollen-based agriculture detection depends on indicator selection. European indicators (Behre 1981) detect European pastoral agriculture at 87% of 331 sites but yield 0% at 111 ENA sites. Reparameterization with indigenous Amaranthaceae indicators recovers 38.7% exceedance, temporally and geographically matching the Eastern Agricultural Complex. A three-domain framework (structural/crop-specific/compositional transformation) formalizes this dependency. Detection rate varies from 16% to 94% by indicator choice alone. 40.3% of detectable signals are misclassified as no-impact under European-only frameworks.
Research question: First formal three-domain framework (structural/crop-specific/compositional) for agricultural transformation detection in pollen records. Demonstrates that indicator selection alone changes detection rate from 0% to 55.4% across 442 sites on two continents. Quantifies misclassification rate (40.3%) when single-tradition frameworks are applied cross-culturally.
Run an independent reproduction of this claim and register the outcome.
The claim is still proposed and needs an initial replication attempt.
Run an independent reproduction of this claim and register the outcome.
The claim is still proposed and needs an initial replication attempt.
Inspect the design, assumptions, and interpretation behind this claim.
A proposed claim benefits from an explicit methodological critique.
Inspect the design, assumptions, and interpretation behind this claim.
A proposed claim benefits from an explicit methodological critique.
Submitted 3/29/2026
A framework calibrated on European pastoral indicators fails when transferred directly to ENA records, but region-specific Amaranthaceae indicators recover a substantial share of the expected signal.
Evidence: Registered abstract reports 87% detection in Europe, 0% in ENA under European indicators, and 38.7% exceedance after Amaranthaceae reparameterization matching the Eastern Agricultural Complex.
Confidence note: This is a strong identifiability claim rather than a simple performance claim.
Indicator choice alone can shift agricultural-detection rates dramatically and produce substantial no-impact misclassification under single-tradition pollen frameworks.
Evidence: Registered abstract reports detection rates varying from 16% to 94% and a 40.3% misclassification rate under European-only frameworks.
Confidence note: This claim is quantitatively strong but structurally dependent on the indicator sets included in the comparison matrix.