Machine-learning prediction over known sites finds earthwork and dark-earth extents "limited, albeit impressive," with other archaeological site types predicted to dominate more than two-thirds of Amazonia.

Machine-learning predictive modeling by Walker et al. (2023) evaluated the spatial distribution of known Amazonian sites and concluded that while the extents of earthworks and anthropogenic dark earths are impressive, they are geographically delimited, with other site types predicted to dominate more than two-thirds of the Amazon basin. This rigorous spatial analysis directly supports the mosaic landscape hypothesis over pan-Amazonian high-density urbanization. The strongest opposing consideration is that machine-learning algorithms trained on an incomplete archaeological catalog inevitably inherit existing geographic and taphonomic discovery biases.

The Walker et al. 2023 quotes — "limited, albeit impressive" predicted earthwork and dark-earth extents, with other site types predicted to dominate more than two-thirds of Amazonia — are verified via the PMC copy. Strongest opposing consideration: the prediction is trained on known sites and inherits discovery bias, and it predates the 2026 LiDAR campaign that multiplied known earthworks several-fold in sampled areas, so its calibration is uncertain in both directions.

No evidence record presents the Walker machine-learning result that earthwork and dark-earth extents were limited or that other site types dominate more than two-thirds of Amazonia. AMZ-E038 and AMZ-E039 concern a different predictive model and cannot substitute for those specific findings. The strongest opposing consideration is thus the complete absence of claim-specific evidence in the evidence ledger.

Walker et al. 2023 is sourced for the “limited, albeit impressive” earthwork and dark-earth extents and dominance of other site types over more than two-thirds of Amazonia, but no evidence record in the provided ledger captures or weighs that result. Scoring remains unresolved under evidence-only rules.

The quoted phrases — "limited, albeit impressive" extents for earthworks and ADEs, and other site types dominating more than two thirds of Amazonia — were verified against the PMC copy, so the claim accurately reports the study. It also converges with AMZ-C029 and AMZ-E010 on the spatial restriction of monumental signatures. The strongest opposing consideration is that this is a machine-learning prediction trained on presence-only records with the same accessibility bias as every other site database, and that "other site types dominate" is itself a claim of widespread non-monumental occupation — so the record cuts against basin-wide earthwork maximalism without supporting emptiness.

Rests on a verified source anchor (Walker et al. 2023) quoting the limited, albeit impressive, predicted extents of earthworks and dark earths and other sites dominating more than two-thirds of Amazonia; no separate evidence record carries it. The model is built from the same biased site base as AMZ-C013.

Held on its anchor alone: the ‘limited, albeit impressive’ phrase and the two-thirds figure are located in the source's text but no evidence record has been admitted. The model is also built from the same access-biased site base as E010.

Held on its anchor alone; no evidence record has been admitted from Walker et al.

Held on its anchor alone: no admitted evidence record from Walker et al. 2023 exists in this ledger.

Machine-learning models of site distributions conclude that the spatial extents of earthworks and anthropogenic dark earths are "limited, albeit impressive," predicting that other, less monumental site types dominate over two-thirds of the Amazon basin (Walker et al. 2023). Strongest opposing consideration: Machine-learning outputs reflect existing training data, which under-sample densely forested or remote interfluvial sectors.

The Walker et al. source is verified with both quoted phrases. The strongest opposing consideration: this is a machine-learning prediction over known sites that inherits the same survey-effort biases as any site database, so "limited" extents are a model output, not an independently surveyed emptiness.

No evidence record reports the machine-learning finding that earthwork and dark-earth extents are “limited, albeit impressive” or that other site types dominate more than two-thirds of Amazonia. Other records support heterogeneity but cannot establish these particular model outputs. The strongest favorable consideration is their consistency with AMZ-E010 and AMZ-E011, which is indirect rather than confirmatory.

Walker et al. 2023’s verified source text describes earthwork and dark- earth predicted extents as “limited, albeit impressive,” with other site types predicted to dominate more than two-thirds of Amazonia. Absence of a dedicated evidence-ledger row slightly reduces formality; the measurement claim is otherwise straightforward and directionally congruent with AMZ-C013/C029.

No evidence record in the packet cites SRC-WALKER-2023; the basis is the claim's sourceAnchor quote plus the source record's verification note, which reproduces both the "limited, albeit impressive" phrase and the "other sites dominate more than two thirds" finding from the PMC copy. I disclose that as the basis. Beyond the thin record, the substantive caution is that a machine-learning prediction trained on presence-only, survey-biased site data is being used to bound the extent of the very site types most likely to be under-detected.

Walker et al. (2023) apply machine learning to archaeological site distributions and conclude that while monumental earthworks and ADEs are impressive, their geographical footprint is restricted, with other lower-intensity site types predicted to dominate over two-thirds of the biome. The strongest opposing consideration is that machine learning predictions are inherently constrained by the presence-only biases of known archaeological catalogs.

The 'limited, albeit impressive' finding and the two-thirds dominance of other site types are verified from the PMC copy of Walker et al. 2023. The opposing consideration is that it is a presence-only ML prediction sharing the survey-bias problems of the whole site-database genre, so it grades the current record rather than measuring the ground truth.

No admitted evidence record reports the machine-learning conclusion that earthwork and dark-earth extents are limited or that other site types dominate more than two-thirds of Amazonia. The strongest consideration against an unresolved verdict is that the source catalog reproduces those phrases, but source metadata is not an evidence record under the governing rules. There is no admitted contrary evidence, so unresolved rather than contradicted is warranted.

No evidence record quotes or links Walker et al. 2023 to this claim; the source is catalogued but not entered as weighed evidence. Under the rules the claim therefore lacks support.

No evidence record is attached. The verdict rests on the claim's anchor and the source record's verification note, which quote 'the limited, albeit impressive, predicted extents of earthworks and ADEs' and 'other sites dominate more than two thirds of all Amazonia'. The claim matches those quotations closely. The strongest opposing consideration is that a machine-learning prediction trained on known sites inherits their survey bias. The second clause also cuts both ways: other site types dominating two thirds of the basin implies widespread if less monumental occupation, not an empty matrix.

Held on its source statement in Walker et al. 2023 ('limited, albeit impressive'), with no checked evidence record taken from it. It is a model trained on survey-biased site data.