CoChat
Sign Up
Cited by 1Amazonian Archaeology and EthnohistoryArchaeological Research and ProtectionArchaeology and ancient environmental studies

The DOI matches its Crossref record; 49 of the 49 references it cites were found in a registry or an index; none of the 80 sources its claims cite is checked for support yet.

Ecological legacies of pre-Columbian settlements evident in palm clusters of neotropical mountain forests

Open access
Fajardo, Sebastian, Mohammadi, Sina, Gregorio de Souza, Jonas, Ardila, César, Tapscott, Alan, Heidgen, Shaddai, Mayorga Hernández, Maria Isabel, Mota de Oliveira, Sylvia, Montejo, Fernando, Moderato, Marco, Peripato, Vinicius, Puche, Katy, Reina, Carlos, Vargas, Juan Carlos, Takes, Frank W., Madella, Marco·2026·Scientific Reports
DOI 10.1038/s41598-026-45976-2PMC PMC13187334
Open this paper in CoChat — free account

Abstract

Ancient populations inhabited and transformed neotropical forests, yet the spatial extent of their ecological influence remains underexplored at high resolution. Here we present a deep learning and remote sensing based approach to estimate areas of pre-Columbian forest modification based on modern vegetation. We apply this method to high-resolution satellite imagery from the Sierra Nevada de Santa Marta, Colombia, as a demonstration of a scalable approach, to evaluate palm tree distributions in relation to archaeological infrastructure. Our findings document a non-random spatial association between archaeological infrastructure and contemporary palm concentrations. Palms were significantly more abundant near archaeological sites with large infrastructure investment. The extent of the largest palm cluster indicates that ancient human-managed areas linked to major infrastructure sites may be up to two orders of magnitude bigger than indicated by current archaeological evidence alone. These patterns are consistent with the hypothesis that past human activity may have influenced local palm abundance and potentially reduced the logistical costs of establishing infrastructure-heavy settlements in less accessible locations. More broadly, our results highlight the utility of palm landscape distributions as an interpretable signal within environmental and multispectral datasets for constraining predictive models of archaeological site locations.

Overview

AI snapshot

Written from the paper’s full text.

TL;DR

A deep-learning and remote-sensing framework (PABAM) applied to the Sierra Nevada de Santa Marta finds non-random spatial association between pre-Columbian archaeological infrastructure and modern palm concentrations, especially around Teyuna. The largest palm cluster suggests ancient human-managed areas may be up to two orders of magnitude larger than current archaeological evidence alone indicates.

Key findings

  • Palms were significantly more abundant near archaeological sites with large infrastructure investment; the largest palm cluster (Cluster 15) spans approximately 100 km², contains ~64,000 palm detections, and includes Teyuna plus 17 archaeological zones.
  • Inverse Distance Weighting scores were much higher at the Teyuna centroid (3.75) than at northwest (0.005), southwest Ciudad Antigüa (0.002), and south-centre (0) centroids, suggesting substantially greater palm density around Teyuna than other possible central sites in highly intervened landscapes.
  • Archaeological centroids in Cluster 15 span 819–1570 m elevation (mean 1113 m); site-associated palms tend to occur at lower elevations than non-site palms and the GBIF baseline, though all detected palm subsets fall within the GBIF elevation envelope.
  • Modern buildings showed only a weak correlation with palm detections (Spearman ρ = 0.21, p < 0.001), and Cluster 15 showed significant spatial aggregation across 96% of evaluated distance bins (r_sig_first = 20 m, r_sig_last = 500 m; L_max = 31.3 at r = 500 m; L_integral = 11174).

Methods

  • PABAM combines YOLOv10x object detection on 30 cm Pléiades Neo panchromatic imagery (2022–2024; 47 tiles from 9 scenes) with archaeological legacy data; a 69.5 km² subset was manually annotated with 24,430 palm trees, then the model was applied across a ~765 km² study area in Colombia.
  • Detected palms were clustered with HDBSCAN after a minimum-cluster-size sweep from 5 to 600, selecting 100; spatial association with archaeological centroids was assessed using IDW scores within a 1000 m radius (w = 1), bootstrapping (1000 samples; 17 archaeological centroids in Cluster 15; 100 random controls), Ripley’s K/L-function analyses, Mann-Whitney U tests, Spearman correlation with modern buildings, and elevation comparisons against GBIF/SRTM data.

Limitations

  • Paper-stated: causal attribution is not definitive; alternative explanations such as preferential settlement in naturally palm-rich areas or shared environmental factors cannot be ruled out, no universal settlement-size threshold is established, and the Cluster 15 interpretation remains speculative and requires ground-truthing.
  • Paper-stated: the model detects only star-shaped canopy crowns, missing understory palms or different crown patterns, and can produce false positives from human-made structures in modified landscapes; GBIF is descriptive rather than a pristine baseline; age–depth models are approximate; some regular cluster classifications may be methodological artefacts; Group A had only n = 2.
  • Evident from the text: archaeological legacy locations carry varying certainty, and western parts of the study area are heavily modern-intervened, limiting direct comparability; the statistical overlap in palm density between archaeological and non-archaeological points within Cluster 15 means the signal is broad and not diagnostic of site centres.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1038/s41598-026-45976-2

Cites

49 of 49 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Verification

Claims the paper makes

0 of 80 cited sources checked · 0 supported · 0 flagged

  • Findingown data

    For the multi-group analysis, the dominant spatial pattern across both groups was aggregation.

    “For the multi-group analysis, the dominant spatial pattern across both groups was aggregation.”
    Quote in the paperNot a substantial claim
  • Findingown data

    Palm density does not differ significantly between archaeological and non-archaeological locations in Cluster 15.

    “This overlap suggests that palm density does not differ significantly between archaeological and non-archaeological locations in this cluster (Fig. 4A).”
    Quote in the paperNot a substantial claim
  • Findingnone given

    Palm detections are significantly more clustered than expected under CSR across virtually all evaluated spatial scales.

    “Palm detections are significantly more clustered than expected under CSR across virtually all evaluated spatial scales.”
    Quote in the paperSubstantial claim
  • Findingown data

    Clusters that spatially incorporate high-investment archaeological sites-most notably Teyuna-exhibit substantially higher palm densities and larger spatial extents than clusters occurring in areas without archaeological occupation.

    “Clusters that spatially incorporate high-investment archaeological sites—most notably Teyuna—exhibit substantially higher palm densities and larger spatial extents than clusters occurring in areas without archaeological occupation.”
    Quote in the paperSubstantial claim
  • Findingown data

    Patterns of certain plant frequencies coincide with nearby archaeological sites.

    “Our data suggest that patterns of certain plant frequencies coincide with nearby archaeological sites.”
    Quote in the paperSubstantial claim
  • Probably not a substantial claim. 45
  • It may state two separate claims as one. 56

  • Probably not a substantial claim. The paper’s conclusions may not depend on it.

    Findingown data

    With a minimum cluster size of five, more than 2000 small clusters were detected.

    “With a minimum cluster size of five, more than 2000 small clusters were detected (Fig. 2A).”
    Quote in the paperNot a substantial claim

  • It may state two separate claims as one.

    Findingown data

    Low minimum cluster size values generate hundreds to thousands of highly localised clusters, many likely spurious, whereas very large values reduce cluster numbers and obscure localised spatial patterns.

    “Our analysis (Fig. 2) shows that low minimum cluster size values generate hundreds to thousands of highly localised clusters, many of which are likely spurious, whereas very large values reduce cluster numbers and obscure localised spatial patterns.”
    Quote in the paperSubstantial claim

  • It may state two separate claims as one.

    Findingown data

    The majority of detected palm clusters are smaller, less dense, and spatially isolated, showing no association with known archaeological features.

    “In contrast, the majority of detected palm clusters are smaller, less dense, and spatially isolated, showing no association with known archaeological features.”
    Quote in the paperSubstantial claim

  • It may state two separate claims as one.

    Findingown data

    A remote sensing analysis estimates the footprint of Teyuna at approximately 100 ha, supporting the role of the settlement as a regional centre within an extensive and interconnected system, but also suggesting a broader anthropogenic landscape surrounding the site.

    “A remote sensing analysis estimates the footprint of Teyuna at approximately 100 ha (Fig. 1), supporting the role of the settlement as a regional centre within an extensive and interconnected system, but also suggesting a broader anthropogenic landscape surrounding the site.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim. The paper’s conclusions may not depend on it.

    Findingnone given

    The spatial clustering pattern within Cluster 15 is qualitatively consistent with the aggregated patterns observed across the majority of smaller HDBSCAN clusters in the multi-group analysis.

    “The spatial clustering pattern within Cluster 15 is qualitatively consistent with the aggregated patterns observed across the majority of smaller HDBSCAN clusters in the multi-group analysis.”
    Quote in the paperNot a substantial claim

  • It may state two separate claims as one.

    Findingown data

    A minimum cluster size of 100 produced stable clusters with high persistence and a noise fraction of approximately 5%, retaining most data points while filtering out low-density groupings.

    “A minimum cluster size of 100 produced a number of stable clusters with high persistence and a noise fraction of approximately 5%, indicating that most data points were retained while filtering out low-density groupings.”
    Quote in the paperSubstantial claim

  • It may state two separate claims as one.

    Findingown data

    The largest archaeological site, Teyuna, is located within the largest cluster of palms, Cluster 15, which spans approximately 100 km^2.

    “Notably, the largest archaeological site (Teyuna) is located within the largest cluster of palms (Cluster 15), spanning approximately 100 km^2 across both well-lit ridges and steep shaded slopes in the eastern part of the study area (Fig. 3).”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.

    Findingnone given

    Some areas in the study region-particularly in the southwest-show signs of modern human intervention, while the area around Teyuna appears more preserved.

    “Some areas in the study region—particularly in the southwest—show signs of modern human intervention, while the area around Teyuna appears more preserved.”
    Quote in the paperNot a substantial claim

  • It may state two separate claims as one.

    Findingnone given

    Palm clustering is common across the study area and reflects a range of ecological processes independent of human activity.

    “Importantly, palm clustering is common across the study area and reflects a range of ecological processes independent of human activity.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim.

    Findingown data

    The noise fraction increased gradually as more points were excluded from larger clusters.

    “The noise fraction (Fig. 2B) increased gradually as more points were excluded from larger clusters.”
    Quote in the paperSubstantial claim

  • It may state two separate claims as one.

    Findingown data

    IDW values for centroids in the northwest, southwest (Ciudad Antigüa), and south-centre regions were 0.005, 0.002, and 0, respectively, while the Teyuna centroid yielded an IDW value of 3.75.

    “The IDW values for the centroids in the northwest, southwest (Ciudad Antigüa), and south-centre regions were 0.005, 0.002, and 0, respectively. In contrast, the Teyuna centroid yielded an IDW value of 3.75.”
    Quote in the paperSubstantial claim

  • It may state two separate claims as one.

    Findingreasoning

    The palm detection model successfully captured the overall distribution and density patterns of palm trees throughout the canopy of the study area, but occasionally failed to accurately detect individual trees.

    “Although the palm detection model used in this study successfully captured the overall distribution and density patterns of palm trees throughout the canopy of the study area, it occasionally failed to accurately detect individual trees.”
    Quote in the paperNot a substantial claim

  • It may state two separate claims as one.

    Findingown data

    Clusters 1, 2, and 15 exhibited spatial overlap with known archaeological features.

    “Clusters 1, 2, and 15, which are located in the southwest, west, and east portions of the study area, respectively (Fig. 3), exhibited spatial overlap with known archaeological features.”
    Quote in the paperSubstantial claim

  • It may state two separate claims as one.

    Findingown data

    Group A, clusters spatially coinciding with known archaeological sites except Cluster 15, showed a mixed pattern, with Cluster 2 classified as clustered and Cluster 1 as regular, though the very small sample size precludes meaningful generalisation.

    “Group A (clusters spatially coinciding with known archaeological sites, except Cluster 15) showed a more mixed pattern, with one cluster classified as clustered (Cluster 2) and one as regular (Cluster 1), though the very small sample size of this group precludes any meaningful generalisation.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.

    Findingreasoning

    Misidentifications mainly occur in heavily modified landscapes; in well-preserved forests-where this approach is intended-the risk of confusing natural features with roads or buildings is minimal.

    “These misidentifications mainly occur in heavily modified landscapes; in well-preserved forests—where this approach is intended—the risk of confusing natural features with roads or buildings is minimal.”
    Quote in the paperNot a substantial claim

  • It may state two separate claims as one.

    Findingown data

    These phytolith assemblages do not include conical forms associated with Dictyocaryum palms, nor do they clearly capture the 1100-1250 CE peak observed in the pollen record.

    “These assemblages do not, however, include conical forms associated with Dictyocaryum palms, nor do they clearly capture the 1100–1250 CE peak observed in the pollen record.”
    Quote in the paperNot a substantial claim

  • It may state two separate claims as one.

    Findingown data

    The spatial association between palm clusters and archaeological sites is not uniform across the landscape; it is spatially variable, with the strongest co-occurrence near the major centre of Teyuna.

    “Our results show that the spatial association between palm clusters and archaeological sites is not uniform across the landscape. Instead, it is spatially variable, with the strongest co-occurrence observed near the major centre of Teyuna.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim. The paper’s conclusions may not depend on it.

    Findingown data

    Cluster 15 contains 17 archaeological zones recorded as points or polygons.

    “This cluster also contains 17 archaeological zones recorded as points or polygons.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim. The paper’s conclusions may not depend on it.

    Findingown data

    Total cluster stability initially declined sharply before stabilizing.

    “Total cluster stability (Fig. 2C) initially declined sharply before stabilizing.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim. The paper’s conclusions may not depend on it.

    Findingreasoning

    Errors in palm detection are primarily caused by visual similarities between palm trees and certain human-made structures, such as crossroads or buildings, that can exhibit textures or shapes resembling the star-like appearance of a palm tree crown.

    “These errors are primarily caused by visual similarities between palm trees and certain human-made structures, such as crossroads or buildings, that can exhibit textures or shapes resembling the star-like appearance of a palm tree crown.”
    Quote in the paperNot a substantial claim

  • It may state two separate claims as one.

    Findingown data

    We document an extensive and dense cluster at the landscape level and natural range distributional shifts towards lower elevations for archaeological site-associated palms.

    “By applying a scalable methodology that links fine-scale plant distributions with archaeological locations, we document an extensive and dense cluster at the landscape level and natural range distributional shifts towards lower elevations for archaeological site-associated palms.”
    Quote in the paperSubstantial claim

  • It may state two separate claims as one.

    Findingown data

    Around a minimum cluster size of 100, the number of clusters stabilized, indicating more robust groupings.

    “Around 100, the number of clusters stabilized, indicating more robust groupings.”
    Quote in the paperSubstantial claim

  • It may state two separate claims as one.

    Findingown data

    Beyond a minimum cluster size of 100, stability changed little, suggesting remaining clusters persist across distance scales.

    “Beyond this value, stability changed little, suggesting that remaining clusters persist across distance scales.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim.

    Findingown data

    Cluster 15 comprised approximately 64000 palm detections distributed across a convex hull of roughly 100 km2.

    “Cluster 15 was analysed separately due to its exceptional size, comprising approximately 64000 palm detections distributed across roughly a convex hull of 100 km².”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim. It may state two separate claims as one.

    Findingnone given

    Two categories based on archaeological association were analysed independently: Group A (clusters 1 and 2, n=2) spatially coinciding with known archaeological sites, and Group B (clusters 3-14 and 16-21, n=18) with no such association.

    “Two categories based on archaeological association were analysed independently: clusters spatially coinciding with known archaeological sites (Group A, n = 2: clusters 1 and 2) and clusters with no such association (Group B, n = 18: clusters 3–14 and 16–21).”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim.

    Premisecitation

    Spatial relationships between plants potentially used in the past for raw materials and ancient settlements of different sizes and spatial arrangements remain poorly understood.

    “While the distribution of tree species exploited for food is relatively well documented18–20, the spatial relationships between plants that could potentially be used in the past for raw materials and ancient settlements of different sizes and spatial arrangements remain poorly understood.”
    Quote in the paperSubstantial claim0 of 3 sources checked
    • cites [18] Iriarte, J. et al. The origins of Amazonian landscapes: Plant cultivation, domestication and the spread of food production in tropical South America. Quat. Sci. Rev. 248, 106582. https://doi.org/10.1016/j.quascirev.2020.106582 (2020). source
    • cites [19] Clement, C. R. et al. Disentangling Domestication from Food Production Systems in the Neotropics. Quaternary https://doi.org/10.3390/quat4010004 (2021). source
    • cites [20] Piperno, D. R. The origins of plant cultivation and domestication in the New World Tropics: Patterns, process, and new developments. Curr. Anthropol. 52(S4), S453–S470 (2011). source

  • It may state two separate claims as one.

    Premisecitation

    Impacts of past activities on modern forest structures near ancient settlements are long-lasting and may influence current estimates of tropical forest composition.

    “These studies show that the impacts of past activities on modern forest structures near the location of ancient settlements are long-lasting, to the extent that they may influence our current estimates of forest composition in tropical forests, such as the Amazon7,8, the eastern flank of the tropical Andes6, and the neotropical forests of Central America9–11.”
    Quote in the paperSubstantial claim0 of 6 sources checked
    • cites [7] McMichael, C. H., Feeley, K. J., Dick, C. W., Piperno, D. R. & Bush, M. B. Comment on “Persistent effects of pre-Columbian plant domestication on Amazonian forest composition”. Science 358(6361), eaan8347. https://doi.org/10.1126/science.aan8347 (2017). source
    • cites [8] McMichael, C. N. H., Matthews-Bird, F., Farfan-Rios, W. & Feeley, K. J. Ancient human disturbances may be skewing our understanding of Amazonian forests. Proc. Natl. Acad. Sci. 114(3), 522–527. https://doi.org/10.1073/pnas.1614577114 (2017). source
    • cites [6] Sales, R. K. et al. Potential distributions of pre-Columbian people in Tropical Andean landscapes. Philos. Trans. R. Soc. B: Biol. Sci. 377(1849), 20200502. https://doi.org/10.1098/rstb.2020.0502 (2022). source
    • cites [9] Brokaw, N. et al. The ancient Maya and the modern forest. Biotropica 57(1), e13370 (2025). source
    • cites [10] Harvey, W. J. et al. The legacy of Pre-Columbian fire on the pine-oak forests of Upland Guatemala. Front. For. Glob. Change 2, 34 (2019). source
    • cites [11] Hightower, J. N., Butterfield, A. C. & Weishampel, J. F. Quantifying ancient Maya land use legacy effects on contemporary rainforest canopy structure. Remote Sens. 6(11), 10716–10732 (2014). source

  • Probably not a substantial claim. The paper’s conclusions may not depend on it.

    Premisecitation

    Archaeological research documents frequent occupation of montane environments in tropical Andean regions, including Colombia.

    “Archaeological research documents frequent occupation of montane environments in tropical Andean regions6, including Colombia12,13.”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites [6] Sales, R. K. et al. Potential distributions of pre-Columbian people in Tropical Andean landscapes. Philos. Trans. R. Soc. B: Biol. Sci. 377(1849), 20200502. https://doi.org/10.1098/rstb.2020.0502 (2022). source
    • cites [12] Langebaek, C.H., ¿ Cuántos eran?¿‘ Dónde estaban?¿‘ Qué les pasó? Poblamiento indígena en la Colombia prehispánica y su transformación después de la Conquista, in Colombia. Preguntas y respuestas sobre su pasado y su presente (Ediciones Uniandes), pp. 27–52 (2010).

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites [13] Botiva Contreras, A. et al. Colombia prehispánica: regiones arqueológicas (Colcultura, Instituto Colombiano de Antropología) (1989).

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Probably not a substantial claim. It may state two separate claims as one.

    Premisecitation

    Ancient human management included cultivation, tending, and dispersal of tree species for food and technological materials, as well as use of fire for forest clearance and enriching nutrient soil composition.

    “Ancient human management included the cultivation, tending, and dispersal of tree species for food and technological materials1,2, as well as the use of fire for forest clearance and enriching nutrient soil composition3,4.”
    Quote in the paperSubstantial claim0 of 4 sources checked
    • cites [1] Levis, C. et al. Persistent effects of pre-Columbian plant domestication on Amazonian forest composition. Science 355(6328), 925–931 (2017). source
    • cites [2] Coelho, S. D. et al. Eighty-four per cent of all Amazonian arboreal plant individuals are useful to humans. PLOS ONE 16(10), e0257875. https://doi.org/10.1371/journal.pone.0257875 (2021). source
    • cites [3] Leal, A., Gassón, R., Behling, H. & Sánchez, F. Human-made fires and forest clearance as evidence for late Holocene landscape domestication in the Orinoco Llanos (Venezuela). Veget. Hist. Archaeobot. 28(5), 545–557. https://doi.org/10.1007/s00334-019-00713-w (2019). source
    • cites [4] Maezumi, S.Y. et al. Fire-human-climate interactions in the Bolivian Amazon rainforest ecotone from the Last Glacial Maximum to late Holocene. Frontiers in Environmental Archaeology 2 (2023), publisher: Frontiers, https://doi.org/10.3389/fearc.2023.1208985, https://www.frontiersin.org/journals/environmental-archaeology/articles/10.3389/fearc.2023.1208985/full. source

  • It may state two separate claims as one.

    Premisecitation

    Palynological analyses in a radius of 6 km from Teyuna, along with phytolith studies at the site, indicate that landscape transformation began between 600 and 900 CE, likely for small-scale cultivation of maize and manioc.

    “Palynological analyses46,60 in a radius of 6 km from Teyuna, along with phytolith studies at the site26, indicate that landscape transformation began between 600 and 900 CE, likely for small-scale cultivation of maize (Zea mays) and manioc (Manihot sp.).”
    Quote in the paperSubstantial claim0 of 3 sources checked
    • cites [46] Herrera, L. F. Buritaca 200: survey of the archaeology and a palynological investigation In (eds van der Hammen, T. & Ruiz, P. M.) (1984).

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites [60] Herrera, L. F. Agricultural activity in the Sierra Nevada de Santa Marta (Colombia): historical perspective In (eds van der Hammen, T. & Ruiz, P. M.) (1984).

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites [26] Giraldo, S. Lords of the Snowy Ranges: Politics, place, and landscape transformation in two Tairona towns in the Sierra Nevada de Santa Marta, Colombia, Phd thesis, University of Chicago, Chicago, Illinois (2010).

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.

    Premisecitation

    Short-term successional dynamics following forest disturbance are relatively well understood for lowland Amazonia, but such processes in neotropical montane environments remain less studied.

    “Although short-term successional dynamics following forest disturbance are relatively well understood for lowland Amazonia44—for instance, the dominance of early successional species depending on disturbance scale—such processes in neotropical montane environments remain less studied6,41.”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites [44] Norden, N. et al. Successional dynamics in Neotropical forests are as uncertain as they are predictable. Proc. Natl. Acad. Sci. U. S. A. 112(26), 8013–8018. https://doi.org/10.1073/pnas.1500403112 (2015). source
    • cites [6] Sales, R. K. et al. Potential distributions of pre-Columbian people in Tropical Andean landscapes. Philos. Trans. R. Soc. B: Biol. Sci. 377(1849), 20200502. https://doi.org/10.1098/rstb.2020.0502 (2022). source
    • cites [41] Bush, M. B. et al. Ecological legacies and recent footprints of the Amazon’s Lost City. Nat. Commun. 16(1), 7408 (2025). source

  • Probably not a substantial claim.

    Interpretationnone given

    We propose a multi-component machine learning method to estimate areas of potential ancient human management using known archaeological sites and vegetation patterns as inputs.

    “We propose a multi-component machine learning method to estimate areas of potential ancient human management using known archaeological sites and vegetation patterns as inputs.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.

    Premisereasoning

    A one-stage object detection model provides a simpler, more reproducible pipeline and enables faster iteration than a multi-component two-stage approach.

    “a one-stage model provides a simpler, more reproducible pipeline and enables faster iteration than a multi-component two-stage approach.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim. It may be too vague to check against evidence. It may state two separate claims as one.

    Interpretationreasoning

    Rather than providing definitive evidence of anthropogenic forest modification, our approach offers a scalable, hypothesis-generating framework for identifying candidate areas where long-term human-environment interactions may have been particularly intense.

    “Rather than providing definitive evidence of anthropogenic forest modification, our approach offers a scalable, hypothesis-generating framework for identifying candidate areas where long-term human–environment interactions may have been particularly intense.”
    Quote in the paperNot a substantial claim

  • It may state two separate claims as one.

    Interpretationreasoning

    The methodology developed here provides a robust framework for detecting subtle ecological signatures of past human activity and can be applied to other neotropical and tropical environments.

    “The methodology developed here, especially in combination with the referred methods, provides a robust framework for detecting subtle ecological signatures of past human activity and can be applied to other neotropical and tropical environments to further refine our understanding of ancient landscape modification.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.

    Premisecitation

    Dictyocaryum lamarckianum-common in the SNSM at elevations of 900-2000 m-is used for construction and may represent one of the sources of phytoliths within the recovered assemblages.

    “Among palms, Dictyocaryum lamarckianum—common in the SNSM at elevations of 900–2000 m—is used for construction and may represent one of the sources of phytoliths within the recovered assemblages26,62.”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [26] Giraldo, S. Lords of the Snowy Ranges: Politics, place, and landscape transformation in two Tairona towns in the Sierra Nevada de Santa Marta, Colombia, Phd thesis, University of Chicago, Chicago, Illinois (2010).

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites [62] van der Hammen, P.M. Thomas; Ruiz, La Sierra Nevada de Santa Marta (Colombia) Transecto Buritaca – La Cumbre, vol. 2 (Schweizerbart Science Publishers, Stuttgart, Germany) (1986), http://www.schweizerbart.de//publications/detail/isbn/9783443650018/Studies_on_Tropical_Andean_Ecosystems_.

  • Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.

    Premisecitation

    For HDBSCAN, small minimum cluster size values produce numerous small clusters, whereas larger values merge smaller groups into fewer, more stable clusters that can hide variability.

    “Small values produce numerous small clusters potentially appearing only at extreme density thresholds, whereas larger values merge smaller groups into fewer, more stable clusters, that can potentially hide the variability of the patterns observed.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [65] Campello, R. J. G. B., Moulavi, D. & Sander, J. Density-Based Clustering Based on Hierarchical Density Estimates In (eds Pei, J. et al.) (2013). source

  • It may state two separate claims as one.

    Interpretationown data

    These patterns are consistent with the hypothesis that past human activity may have influenced local palm abundance and potentially reduced the logistical costs of establishing infrastructure-heavy settlements in less accessible locations.

    “These patterns are consistent with the hypothesis that past human activity may have influenced local palm abundance and potentially reduced the logistical costs of establishing infrastructure-heavy settlements in less accessible locations.”
    Quote in the paperSubstantial claim

  • It may state two separate claims as one.

    Premiseown data

    Pollen data showed a pronounced peak in undifferentiated palm types during 1100-1250 CE, followed by a second, much less pronounced peak around 1650 CE, suggesting that palm abundance increased episodically rather than following a linear long-term trend.

    “Specifically, pollen data showed a pronounced peak in undifferentiated palm types during this interval, followed by a second, much less pronounced peak around 1650 CE, suggesting that palm abundance increased episodically rather than following a linear long-term trend.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim. The paper’s conclusions may not depend on it.

    Premisecitation

    Teyuna has been occupied since around 400 CE.

    “Although Teyuna has been occupied since around 400 CE26”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [26] Giraldo, S. Lords of the Snowy Ranges: Politics, place, and landscape transformation in two Tairona towns in the Sierra Nevada de Santa Marta, Colombia, Phd thesis, University of Chicago, Chicago, Illinois (2010).

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Probably not a substantial claim. The paper’s conclusions may not depend on it.

    Premisecitation

    Bush et al.'s approach favours a combined explanation involving indirect human legacy and climate-driven fluctuations for the increased abundance of Dictyocaryum.

    “Their approach, using different methods and data sources, favours a combined explanation involving indirect human legacy and climate-driven fluctuations for the increased abundance of Dictyocaryum.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Bush et al. [41] Bush, M. B. et al. Ecological legacies and recent footprints of the Amazon’s Lost City. Nat. Commun. 16(1), 7408 (2025). source

  • It may state two separate claims as one.

    Premisecitation

    The composition of the SNSM forest in areas without reported human presence indicates that palms are a key component of the floristic assemblage but their frequency in the vegetation is relatively low.

    “The composition of the SNSM forest in areas without reported human presence indicates that palms are a key component of the floristic assemblage but their frequency in the vegetation is relatively low42,43.”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [42] Fortier, R. P. et al. Diversity, composition, and structure of a 1-hectare tree plot in the cloud forest of the Sierra Nevada de Santa Marta. Colombia. Ann. For. Sci. 81(1), 24. https://doi.org/10.1186/s13595-024-01239-6 (2024). source
    • cites [43] Dechner, A. & Diazgranados, M. Floristic composition and structure of forested areas along the low basin of the San Salvador river, north side of Sierra Nevada de Santa Marta. Colombia. Univ. Sci. 12(2), 99–124 (2007).

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • It may state two separate claims as one.

    Study design and conductreasoning

    Detections are inherently biased toward crowns exposed at or near the canopy surface; palms located in the understory or obscured by overstory vegetation are less likely to be detected.

    “Since orbital optical imagery captures reflected radiance from the uppermost visible surfaces (here, the panchromatic band), detections are inherently biased toward crowns exposed at/near the canopy surface; palms located in the understory or obscured by overstory vegetation are less likely to be detected.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim. The paper may give no reason why it holds.

    Premisenone given

    Traditional field-based approaches are difficult to apply across steep, densely forested montane terrain.

    “However, traditional field-based approaches are difficult to apply across steep, densely forested montane terrain.”
    Quote in the paperNot a substantial claim

  • It may state two separate claims as one.

    Premisecitation

    There is no uncertainty that montane regions were occupied; archaeological evidence demonstrates sustained pre-Columbian presence.

    “There is no uncertainty about whether montane regions were occupied—archaeological evidence clearly demonstrates sustained pre-Columbian presence6,12,13—but rather we need to measure how localised human activities may have shaped forest composition and structure across spatially continuous landscapes.”
    Quote in the paperSubstantial claim0 of 3 sources checked
    • cites [6] Sales, R. K. et al. Potential distributions of pre-Columbian people in Tropical Andean landscapes. Philos. Trans. R. Soc. B: Biol. Sci. 377(1849), 20200502. https://doi.org/10.1098/rstb.2020.0502 (2022). source
    • cites [12] Langebaek, C.H., ¿ Cuántos eran?¿‘ Dónde estaban?¿‘ Qué les pasó? Poblamiento indígena en la Colombia prehispánica y su transformación después de la Conquista, in Colombia. Preguntas y respuestas sobre su pasado y su presente (Ediciones Uniandes), pp. 27–52 (2010).

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites [13] Botiva Contreras, A. et al. Colombia prehispánica: regiones arqueológicas (Colcultura, Instituto Colombiano de Antropología) (1989).

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Probably not a substantial claim. It may be too vague to check against evidence.

    Premisereasoning

    Understanding how ancient human activities influenced neotropical montane forests requires analyses covering spatially continuous areas large enough to capture both ecological and archaeological variability.

    “Understanding how ancient human activities influenced neotropical montane forests requires analyses that cover spatially continuous areas large enough to capture both ecological and archaeological variability.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim. It may be too vague to check against evidence.

    Interpretationnone given

    Our findings highlight the challenges inherent in attributing modern vegetation patterns to ancient human activities in complex tropical landscapes.

    “Our findings highlight the challenges inherent in attributing modern vegetation patterns to ancient human activities in complex tropical landscapes.”
    Quote in the paperNot a substantial claim

  • It may state two separate claims as one.

    Interpretationown data

    While natural ecological processes cannot be entirely ruled out, the density and extent of palms around settlements with engineered landscape features, and paleoecological evidence suggest that the presence of ancient settlements may have influenced these plant communities over long timescales.

    “While natural ecological processes cannot be entirely ruled out, the density and extent of palms around settlements with engineered landscape features, and paleoecological evidence suggest that the presence of ancient settlements may have influenced these plant communities over long timescales.”
    Quote in the paperSubstantial claim

  • It may state two separate claims as one.

    Premisecitation

    Bush et al. found that modern distributions of Dictyocaryum in the northern Andes reflect long-term human influence, showing that contemporary palm locations can carry the signature of past activities.

    “They found that modern distributions of Dictyocaryum in the northern Andes reflect long-term human influence, showing that contemporary palm locations can carry the signature of past activities.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Bush et al. [41] Bush, M. B. et al. Ecological legacies and recent footprints of the Amazon’s Lost City. Nat. Commun. 16(1), 7408 (2025). source

  • It may state two separate claims as one.

    Interpretationcitation

    In the montane forest of the study area, the most likely palm detected under these conditions is Dictyocaryum sp. because it is known for its tall, solitary growth form and prominent canopy presence in Andean cloud forests.

    “In the montane forest of the study area, the most likely palm detected under these conditions is Dictyocaryum sp. This genus is known for its tall, solitary growth form and prominent canopy presence in Andean cloud forests22,63.”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites [22] Paniagua-Zambrana, N. Y. & Bussmann, R. W. Dictyocaryum lamarckianum (Mart.) H. Wendl. In (eds Paniagua-Zambrana, N. Y. & Bussmann, R. W.) (2020). source
    • cites [63] Henderson, A., Galeano, G. & Bernal, R. Field guide to the palms of the Americas (Princeton University Press) (1995).

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • It may state two separate claims as one.

    Premisecitation

    The YOLOv10 extra-large configuration has substantially narrowed the historical accuracy gap with two-stage detectors and performs well on small, visually complex targets such as star-shaped palm crowns.

    “Recent advances in the YOLO family—particularly the v10 extra-large configuration—have substantially narrowed the historical accuracy gap with two-stage detectors and perform well on small, visually complex targets such as star-shaped palm crowns.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [64] Wang, A. et al. YOLOv10: Real-Time End-to-End Object Detection, in Advances in Neural Information Processing Systems, A. Globerson, et al., Eds. (Curran Associates, Inc.), vol. 37, pp. 107984–108011, https://proceedings.neurips.cc/paper_files/paper/2024/file/c34ddd05eb089991f06f3c5dc36836e0-Paper-Conference.pdf (2024). source

  • It may state two separate claims as one.

    Interpretationreasoning

    These clusters likely formed through natural processes and serve as a useful baseline against which the archaeology-linked clusters can be evaluated.

    “These clusters likely formed through natural processes and serve as a useful baseline against which the archaeology-linked clusters can be evaluated.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim.

    Premisereasoning

    Automated remote-sensing approaches provide a scalable way to examine relationships between forest composition and archaeological evidence across large and otherwise inaccessible regions.

    “Automated remote-sensing approaches, such as the method used here, provide a scalable way to examine relationships between forest composition and archaeological evidence across large and otherwise inaccessible regions.”
    Quote in the paperSubstantial claim

  • It may state two separate claims as one.

    Premisecitation

    Human occupation in the SNSM is dated from 200 to 1600 CE and is divided into Neguanje (200-1100 CE) and Tairona (1100-1600 CE) periods.

    “Human occupation, dated from 200 to 1600 CE24,26, is divided into two archaeological periods: Neguanje (200–1100 CE) and Tairona (1100–1600 CE).”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [24] Rodríguez Osorio, D., Giraldo, S., Mazuera, E., Burbano, A. & Figueredo, E. Beyond visualization: Remote sensing applications in prehispanic settlements to understand ancient anthropogenic land use and occupation in the Sierra Nevada de Santa Marta. Colombia. Lat. Am. Antiq. 34(4), 821–841 (2023). source
    • cites [26] Giraldo, S. Lords of the Snowy Ranges: Politics, place, and landscape transformation in two Tairona towns in the Sierra Nevada de Santa Marta, Colombia, Phd thesis, University of Chicago, Chicago, Illinois (2010).

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Probably not a substantial claim. The paper may give no reason why it holds.

    Interpretationnone given

    The approach is scalable to any large area with submetric resolution satellite imagery available.

    “The approach is scalable to any large area with submetric resolution satellite imagery available.”
    Quote in the paperNot a substantial claim

  • It may state two separate claims as one.

    Interpretationreasoning

    Contemporary land-use intensity in those areas likely alters or masks palm distribution patterns, limiting direct comparability.

    “The intensity of contemporary land-use in those areas likely alters or masks palm distribution patterns, limiting direct comparability.”
    Quote in the paperSubstantial claim

  • It may state two separate claims as one.

    Study design and conductnone given

    The computer vision model detected palms with a specific crown pattern visible at the top of the canopy; palms with different crown patterns or palms whose crowns were located lower in the canopy were unlikely to be detected by the method.

    “The computer vision model detected palms with a specific crown pattern visible at the top of the canopy. Palms with different crown patterns or palms whose crowns were located lower in the canopy were unlikely to be detected by our method.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim.

    Study design and conductnone given

    The final HDBSCAN minimum cluster size m was selected by balancing cluster stability with removal of trivial clusters while maintaining an acceptable noise fraction.

    “The final m value was selected by balancing cluster stability with the removal of trivial clusters, while maintaining an acceptable noise fraction.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.

    Premisecitation

    Analogous processes of human environmental modification have been documented for animal species in the Late Pleistocene and for plant communities in the construction of artificial forest islands in Amazonian savannahs and in the deliberate enrichment of palm species in Amazonian forests.

    “analogous processes have been documented for animal species in the Late Pleistocene37, and, more relevantly for plant communities, in the construction of approximately 4700 artificial forest islands within seasonally flooded Amazonian savannahs during the Early Holocene38”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites [37] Baumann, C. The paleo-synanthropic niche: A first attempt to define animal’s adaptation to a human-made micro-environment in the Late Pleistocene. Archaeol. Anthropol. Sci. 15, 63. https://doi.org/10.1007/s12520-023-01764-x (2023). source
    • cites [38] Lombardo, U. et al. Early Holocene crop cultivation and landscape modification in Amazonia. Nature 581(7807), 190–193. https://doi.org/10.1038/s41586-020-2162-7 (2020). source
    • cites [18] Iriarte, J. et al. The origins of Amazonian landscapes: Plant cultivation, domestication and the spread of food production in tropical South America. Quat. Sci. Rev. 248, 106582. https://doi.org/10.1016/j.quascirev.2020.106582 (2020). source

  • It may state two separate claims as one.

    Premisecitation

    The largest known settlement is Teyuna, occupied in the Neguanje period from around 400 CE; during the Tairona period it expanded to 33 ha.

    “The largest known settlement is Teyuna (also known as Ciudad Perdida or Buritaca-200), occupied in the Neguanje period from around 400 CE26. During the Tairona period, it expanded to a settlement of 33 ha with typical regional architecture24,26,61.”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites [26] Giraldo, S. Lords of the Snowy Ranges: Politics, place, and landscape transformation in two Tairona towns in the Sierra Nevada de Santa Marta, Colombia, Phd thesis, University of Chicago, Chicago, Illinois (2010).

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites [24] Rodríguez Osorio, D., Giraldo, S., Mazuera, E., Burbano, A. & Figueredo, E. Beyond visualization: Remote sensing applications in prehispanic settlements to understand ancient anthropogenic land use and occupation in the Sierra Nevada de Santa Marta. Colombia. Lat. Am. Antiq. 34(4), 821–841 (2023). source
    • cites [61] Mayorga Hernández, M.I. El agua como elemento constructor de territorio y arquitectura en Colombia. Caso de estudio Teyuna, Sierra Nevada de Santa Marta, Phd thesis, Universidad de Granada, Granada, Spain (2024), https://hdl.handle.net/10481/89459.

  • It may state two separate claims as one.

    Premisecitation

    Tairona sites feature terraced platforms with curved stone walls, rammed earth foundations, rainwater drainage systems, and circular stone floors for wood-and-thatch structures connected by paved paths.

    “Spanning from coastal lowlands to 2000 m elevation, the Tairona sites feature terraced platforms with curved stone walls, rammed earth foundations, rainwater drainage systems, and circular stone floors for wood-and-thatch structures connected by paved paths26,27,54,59–61.”
    Quote in the paperNot a substantial claim0 of 6 sources checked
    • cites [26] Giraldo, S. Lords of the Snowy Ranges: Politics, place, and landscape transformation in two Tairona towns in the Sierra Nevada de Santa Marta, Colombia, Phd thesis, University of Chicago, Chicago, Illinois (2010).

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites [27] Groot, A. M. Arqueología y conservación de la localidad precolombina de Buritaca 200 en la Sierra Nevada de Santa Marta. Inform. Antropol. 1, 55–102 (1985).

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites [54] Serje, M. Arquitectura y urbanismo en la cultura Tairona. Bol. Mus. Oro 19(1), 87–96 (1987).

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites [59] Cadavid, G. & Herrera, L. F. Manifestaciones culturales en el área Tairona. Inf. Antropol. 1, 5–54 (1985).

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites [60] Herrera, L. F. Agricultural activity in the Sierra Nevada de Santa Marta (Colombia): historical perspective In (eds van der Hammen, T. & Ruiz, P. M.) (1984).

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites [61] Mayorga Hernández, M.I. El agua como elemento constructor de territorio y arquitectura en Colombia. Caso de estudio Teyuna, Sierra Nevada de Santa Marta, Phd thesis, Universidad de Granada, Granada, Spain (2024), https://hdl.handle.net/10481/89459.

  • Probably not a substantial claim. The paper’s conclusions may not depend on it.

    Study design and conductnone given

    Within the training and validation region, 95% of patches were used for training and 5% for validation to guide hyperparameter tuning and model selection.

    “Within the training and validation region, 95% of the patches were used for training and the remaining 5% were used for validation to guide hyperparameter tuning and model selection.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.

    Premisecitation

    Settlements occur on ridgelines, valley slopes, riverbanks, bays, inlets, flat coastal areas, and deeper canyons.

    “Settlements occur on ridgelines, valley slopes24,26,54, riverbanks, bays, inlets, flat coastal areas, and deeper canyons26,55.”
    Quote in the paperNot a substantial claim0 of 4 sources checked
    • cites [24] Rodríguez Osorio, D., Giraldo, S., Mazuera, E., Burbano, A. & Figueredo, E. Beyond visualization: Remote sensing applications in prehispanic settlements to understand ancient anthropogenic land use and occupation in the Sierra Nevada de Santa Marta. Colombia. Lat. Am. Antiq. 34(4), 821–841 (2023). source
    • cites [26] Giraldo, S. Lords of the Snowy Ranges: Politics, place, and landscape transformation in two Tairona towns in the Sierra Nevada de Santa Marta, Colombia, Phd thesis, University of Chicago, Chicago, Illinois (2010).

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites [54] Serje, M. Arquitectura y urbanismo en la cultura Tairona. Bol. Mus. Oro 19(1), 87–96 (1987).

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites [55] Dever, A. Social and economic development of a specialized community in Chengue, Parque Tairona, Colombia, Ph.D. thesis, University of Pittsburgh, Pittsburgh, Pennsylvania (2007).

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • It may state two separate claims as one.

    Premisecitation

    Palynological evidence suggests rapid forest regeneration after site abandonment in the 16th century, implying that Tairona settlements had an ecological footprint on modern forest composition.

    “Palynological evidence also suggests rapid forest regeneration after site abandonment in the 16th century, implying that Tairona settlements had an ecological footprint on modern forest composition60.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [60] Herrera, L. F. Agricultural activity in the Sierra Nevada de Santa Marta (Colombia): historical perspective In (eds van der Hammen, T. & Ruiz, P. M.) (1984).

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • It may state two separate claims as one.

    Premisecitation

    At Teyuna, Dictyocaryum lamarckianum and Arecaceae pollen were detected, coinciding with the first centuries (1100-1250 CE) of the Tairona period.

    “At Teyuna, Dictyocaryum lamarckianum22 and Arecaceae pollen were detected46, coinciding with the first centuries (1100-1250 CE) of the Tairona period, a time of major infrastructural investment in the SNSM.”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites [22] Paniagua-Zambrana, N. Y. & Bussmann, R. W. Dictyocaryum lamarckianum (Mart.) H. Wendl. In (eds Paniagua-Zambrana, N. Y. & Bussmann, R. W.) (2020). source
    • cites [46] Herrera, L. F. Buritaca 200: survey of the archaeology and a palynological investigation In (eds van der Hammen, T. & Ruiz, P. M.) (1984).

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Probably not a substantial claim. The paper’s conclusions may not depend on it.

    Interpretationreasoning

    The exceptional size and density of Cluster 15 may be at least partly attributable to the presence of archaeological sites within its extent, where repeated or sustained human occupation could have promoted palm establishment, dispersal, or maintenance over time.

    “We tentatively suggest that the exceptional size and density of this cluster may be at least partly attributable to the presence of archaeological sites within its extent, where repeated or sustained human occupation could have promoted palm establishment, dispersal, or maintenance over time.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim. It may be too vague to check against evidence. It may state two separate claims as one.

    Interpretationreasoning

    Our approach provides a robust framework for detecting subtle ecological legacies of past human activity and highlights the potential of integrating remote sensing, species detection, and archaeological data to uncover long-term human-environment interactions in neotropical montane forests.

    “Our approach provides a robust framework for detecting subtle ecological legacies of past human activity and highlights the potential of integrating remote sensing, species detection, and archaeological data to uncover long-term human-environment interactions in neotropical montane forests.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.

    Study design and conductown data

    The age-depth models underlying these reconstructions are approximate, relying on interpolated dates from other cores and excavations, so the chronology of these peaks should be interpreted cautiously.

    “However, the age–depth models underlying these reconstructions are approximate, relying on interpolated dates from other cores and excavations, and thus the chronology of these peaks should be interpreted cautiously.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim.

    Study design and conductreasoning

    The analysis focused on the less-disturbed eastern part of the study area to avoid areas with more modern human intervention.

    “We chose to focus our analysis in the less-disturbed eastern part of the area of study because the areas in the western portion of the study area have undergone more modern human intervention.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The western portion of the study area has undergone more modern human intervention than the eastern portion.

  • Probably not a substantial claim. The paper’s conclusions may not depend on it.

    Study design and conductown data

    The consistency of the palm detection model was evaluated by comparing the elevation of detected palms with the elevation distribution of Dictyocaryum occurrences recorded in GBIF.

    “We further evaluated the consistency of the palm detection model by comparing the elevation of detected palms with the elevation distribution of Dictyocaryum occurrences recorded in the Global Biodiversity Information Facility (GBIF) database69.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [69] GBIF, Dictyocaryum H.Wendl. Occurrence Download, https://www.gbif.org/occurrence/download/0025145-250310093411724 (2025), https://doi.org/10.15468/DL.ND44PT, https://www.gbif.org/occurrence/download/0025145-250310093411724, global Biodiversity Information Facility. Accessed: 2025-03-24. source

  • Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.

    Study design and conductreasoning

    The total subsample size for Cluster 15 was set to 5% of the full cluster count, up to a maximum of 10000 points, generating n=3202.

    “The total subsample size was set to 5% of the full cluster count, up to a maximum of 10000 points, generating a substantially higher sampling fraction (n=3202) than the total detected palms of other clusters.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim. It may state two separate claims as one.

    Study design and conductown data

    Bootstrapping was used to compare palm concentrations around archaeological centroids with control zones, with 1000 bootstrapped datasets each of 17 observations sampled with replacement, matching the number of archaeological centroids in the largest cluster.

    “For both control sets and the archaeological centroids within the largest palm cluster, we created 1000 bootstrapped datasets. Each dataset consisted of 17 observations sampled with replacement, matching the number of archaeological centroids in the largest cluster.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.

    Study design and conductreasoning

    GBIF occurrence records were used as an observation-based reference for the species' elevational envelope, but because GBIF data can reflect uneven sampling effort and anthropogenic influences, the comparison is interpreted as descriptive rather than a definitive estimate of a pristine natural range.

    “GBIF occurrence records were used as an observation-based reference for the species’ reported elevational envelope; however, because GBIF data can reflect uneven sampling effort and potential anthropogenic influences, we interpret this comparison as descriptive rather than as a definitive estimate of a pristine natural range.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [69] GBIF, Dictyocaryum H.Wendl. Occurrence Download, https://www.gbif.org/occurrence/download/0025145-250310093411724 (2025), https://doi.org/10.15468/DL.ND44PT, https://www.gbif.org/occurrence/download/0025145-250310093411724, global Biodiversity Information Facility. Accessed: 2025-03-24. source

  • Probably not a substantial claim.

    Study design and conductreasoning

    Palm detection coordinates were projected from geographic coordinates to UTM Zone 18N to ensure all distance calculations were performed in metres.

    “For the Ripley’s K analyses palm detection coordinates were projected from geographic coordinates to UTM Zone 18N to ensure all distance calculations were performed in metres.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.

    Study design and conductreasoning

    Due to differences in sampling design, window area, and r_max scale, Cluster 15 results are reported as a standalone characterisation and are not included in comparative statistical tests between archaeological groups.

    “Given the differences in sampling design, window area, and r_max scale, the Cluster 15 results are reported as a standalone characterisation of within-cluster spatial structure and are not included in the comparative statistical tests between archaeological groups.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim.

    Study design and conductreasoning

    The maximum search radius for Cluster 15 was fixed at r_max = 500 m to assess clustering at fine spatial scales consistent with ecological processes driving palm aggregation, rather than at landscape scales where the large window area would dilute the clustering signal.

    “The maximum search radius was fixed at r_max = 500 m. This choice was deliberate to assess clustering at fine spatial scales consistent with the ecological processes driving palm aggregation, rather than at landscape scales where the large window area would dilute the clustering signal.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim. The paper’s conclusions may not depend on it.

    Study design and conductnone given

    An interquartile range-based outlier filtering on SRTM elevation was applied to the GBIF baseline.

    “An interquartile range-based outlier filtering on the SRTM elevation was applied to the GBIF baseline.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.

    Study design and conductown data

    A 95% pointwise simulation envelope was constructed from 199 Monte Carlo CSR realisations, with simulations seeded by 199 independent child seeds from a single root seed (42) to ensure reproducibility and statistical independence.

    “A 95% pointwise simulation envelope was constructed from 199 Monte Carlo CSR realisations. Simulations were seeded by spawning 199 independent child seeds from a single root seed (42) using np.random.SeedSequence, ensuring both reproducibility and statistical independence between simulations.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim. It may state two separate claims as one.

    Study design and conductcitation

    Ripley's K-function with isotropic edge correction was computed following Ripley, and the centred Besag's L-function transformation was applied.

    “For both protocols, Ripley’s Kfunction with isotropic edge correction was computed following Ripley66,67, and the centred Besag’s L-function transformation was applied68.”
    Quote in the paperSubstantial claim0 of 3 sources checked
    • cites [66] Ripley, B. D. The second-order analysis of stationary point processes. J. Appl. Probab. 13(2), 255–266 (1976). source
    • cites [67] Ripley, B. D. Statistical inference for spatial processes (Cambridge university press, 1988). source
    • cites [68] Besag, J. Contribution to the discussion on Dr Ripley’s paper. JR Stat Soc B 39, 193–195 (1977).

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • It may state two separate claims as one.

    Interpretationreasoning

    Applying simple random subsampling and a fixed window approach to Cluster 15 would have yielded a point density too low for meaningful spatial statistics, because the CSR null model would have been evaluated against a sparsely sampled representation of a large landscape rather than local-scale spatial structure.

    “Applying the simple random subsampling and fixed window approach used in the multi-group analysis would have yielded a point density far too low relative to the window area for meaningful spatial statistics: the CSR null model would have been evaluated against a sparsely sampled representation of a large landscape rather than against the local-scale spatial structure of ecological interest.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.

    Study design and conductreasoning

    Two independent analytical protocols were applied: a multi-group analysis covering all HDBSCAN clusters except Cluster 15, and a dedicated analysis of Cluster 15 due to its exceptional spatial extent.

    “Two independent analytical protocols were applied: a multi-group analysis covering all HDBSCAN clusters (except Cluster 15) and a dedicated analysis of Cluster 15, which required a distinct methodological treatment due to its exceptional spatial extent.”
    Quote in the paperNot a substantial claim
81 of 81

  • The paper relies on this without support.

    In the SNSM, lower elevations are correlated with increased taxa competition for solar radiation due to hilly terrain and persistently cloudy conditions.

    “In the SNSM lower elevations are correlated with increased taxa competition for solar radiation due to the hilly terrain and persistently cloudy conditions.”

  • The paper relies on this without support.

    The western portion of the study area has undergone more modern human intervention than the eastern portion.

    “the areas in the western portion of the study area have undergone more modern human intervention.”
2 of 2

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

  • Four additional centroids are announced, but only three are accounted for

    Minor

    The text says four additional archaeological centroids were included, then identifies one known-site centroid and the remaining two. That accounts for only three.

  • Species-level detection conflicts with the stated detection target

    Minor

    The paper describes the detector as recognizing a crown shape rather than identifying individual palm species, but later characterizes the framework as using species-level detection.

2 of 2

Citations

Cites

Flagged first, then by citations

Research works printed but not linked: 16

  1. Langebaek, C.H., ¿ Cuántos eran?¿‘ Dónde estaban?¿‘ Qué les pasó? Poblamiento indígena en la Colombia prehispánica y su transformación después de la Conquista, in Colombia. Preguntas y respuestas sobre su pasado y su presente (Ediciones Uniandes), pp. 27–52 (2010).
  2. Instituto Geográfico Agustín Codazzi (IGAC). Estudio general de suelos y zonificación de tierras del Departamento de Magdalena (Imprenta Nacional de Colombia, 2009).
  3. Giraldo, S. Lords of the Snowy Ranges: Politics, place, and landscape transformation in two Tairona towns in the Sierra Nevada de Santa Marta, Colombia, Phd thesis, University of Chicago, Chicago, Illinois (2010).
  4. Groot, A. M. Arqueología y conservación de la localidad precolombina de Buritaca 200 en la Sierra Nevada de Santa Marta. Inform. Antropol. 1, 55–102 (1985).
  5. Harvey, W. J. et al. The legacy of pre-Columbian fire on the pine-oak forests of Upland Guatemala. Front. For. Glob. Change 2, 34. https://doi.org/10.3389/ffgc.2019.00034 (2019). · DOI · Paper page
  6. Hightower, J. N., Butterfield, A. C. & Weishampel, J. F. Quantifying ancient Maya land use legacy effects on contemporary rainforest canopy structure. Remote Sens. 6(11), 10716–10732. https://doi.org/10.3390/rs61110716 (2014). · DOI · Paper page
  7. Dechner, A. & Diazgranados, M. Floristic composition and structure of forested areas along the low basin of the San Salvador river, north side of Sierra Nevada de Santa Marta. Colombia. Univ. Sci. 12(2), 99–124 (2007).
  8. Zhang, S., Tuo, H., Hu, J. & Jing, Z. Domain adaptive YOLO for one-stage cross-domain detection, in Asian conference on machine learning pp. 785–797 (PMLR, 2021).
  9. Serje, M. Arquitectura y urbanismo en la cultura Tairona. Bol. Mus. Oro 19(1), 87–96 (1987).
  10. Dever, A. Social and economic development of a specialized community in Chengue, Parque Tairona, Colombia, Ph.D. thesis, University of Pittsburgh, Pittsburgh, Pennsylvania (2007).
  11. Yépez, Y. & Gutiérrez, Y. Relato del espacio, territorio y lugar de los asentamientos Kogui dentro de la Línea Negra: Desarrollo cartográfico de ocupación e integración, Ba thesis (Univerisidad de la Costa C.U.C, Barranquilla, Atlántico, 2020).
  12. Cadavid, G. & Herrera, L. F. Manifestaciones culturales en el área Tairona. Inf. Antropol. 1, 5–54 (1985).
  13. Mayorga Hernández, M.I. El agua como elemento constructor de territorio y arquitectura en Colombia. Caso de estudio Teyuna, Sierra Nevada de Santa Marta, Phd thesis, Universidad de Granada, Granada, Spain (2024), https://hdl.handle.net/10481/89459.
  14. van der Hammen, P.M. Thomas; Ruiz, La Sierra Nevada de Santa Marta (Colombia) Transecto Buritaca – La Cumbre, vol. 2 (Schweizerbart Science Publishers, Stuttgart, Germany) (1986), http://www.schweizerbart.de//publications/detail/isbn/9783443650018/Studies_on_Tropical_Andean_Ecosystems_.
  15. Henderson, A., Galeano, G. & Bernal, R. Field guide to the palms of the Americas (Princeton University Press) (1995).
  16. Besag, J. Contribution to the discussion on Dr Ripley’s paper. JR Stat Soc B 39, 193–195 (1977).

Cited by

Sorted by citations

Related

Metadata

Authors
Fajardo, S., Mohammadi, S., Gregorio de Souza, J., Ardila, C., Tapscott, A., Heidgen, S., Mayorga Hernández, M. I., Mota de Oliveira, S., Montejo, F., Moderato, M., Peripato, V., Puche, K., Reina, C., Vargas, J. C., Takes, F. W., & Madella, M.
Affiliations
Leiden University (NL)
Corresponding author
Fajardo, Sebastian
Venue
Scientific Reports
Publisher
Springer Science and Business Media LLC
Published
2026-04-01
Volume
16
Issue
1
ISSN
2045-2322 (ISSN-L)
DOI
10.1038/s41598-026-45976-2
PMC
PMC13187334
OpenAlex
W6929297954
Cited by
1
References
48
Citations by year
2026: 1
FWCI
57.47
Type
journal article
Language
en
Open access
Licence
Locations
  • Scientific Reports (publishedVersion, cc-by-nc-nd)
  • PubMed Central (publishedVersion, cc-by-nc-nd)
  • PubMed (publishedVersion)
  • arXiv (Cornell University) (submittedVersion)
  • The Digital Academic Repository of Naturalis Biodiversity Center (Naturalis Biodiversity Center) (submittedVersion, cc-by-nc-nd)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
  • Repositori digital de la UPF (Universitat Pompeu Fabra) (submittedVersion, cc-by-nc-nd)
  • Biblioteca Digital da Memória Científica do INPE (National Institute for Space Research) (submittedVersion)
  • arXiv (Cornell University)
Funders
  • Arcadia Fund
  • Leiden Delft Erasmus Global (LDE-127)
  • European Space Agency
Topics
  • Social Sciences › Arts and Humanities › History › Amazonian Archaeology and Ethnohistory
  • Social Sciences › Archeology › Archaeological Research and Protection
  • Social Sciences › Archeology › Archaeology and ancient environmental studies
Keywords
pre-Columbian settlement, ecological legacies, palm distribution, human-modified forests, Sierra Nevada de Santa Marta, deep learning, satellite imagery, landscape archaeology
MeSH terms
Archaeology, Colombia, Humans, Trees, Ecosystem, Arecaceae, Satellite Imagery, Forests

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

A language model drew the claims from the paper on 2026-11-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-11-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review