CoChat
Sign Up
Cited by 7Wildlife Ecology and ConservationAnimal Ecology and Behavior StudiesEcology and Vegetation Dynamics Studies

Non‐random sampling measures the occurrence but not the strength of a textbook trophic cascade

Open access
MacNulty, Daniel R., Brice, Elaine M., Larsen, Eric J.·2024·Ecology Letters
DOI 10.1111/ele.14344
Open this paper in CoChat — free accountOpen PDF

Abstract

Although sampling the five tallest young aspen in a stand is useful for detecting the occurrence of any aspen recruitment, this technique overestimates the population response of aspen to wolf reintroduction. Our original conclusion that random sampling described a trophic cascade that was weaker than the one described by non‐random sampling is unchanged.

AI snapshot

Written from the paper’s full text.

TL;DR

This reply argues that sampling only the five tallest young aspen can reveal whether a wolf–elk–aspen trophic cascade occurred, but it cannot reliably measure the cascade’s strength because it exaggerates population-level aspen responses. The authors maintain that random sampling shows a weaker aspen response to wolf reintroduction than non-random sampling does.

Key findings

  • The five-tallest (5T) approach overestimated overstory aspen regeneration by four- to sevenfold compared with random sampling, according to the authors’ earlier work, so it inflates estimates of the aspen population response to wolf reintroduction.
  • Painter et al. conflate detecting the occurrence of a trophic cascade with measuring its strength; evidence that young aspen heights increased does not establish how strong the wolf–elk–aspen cascade was.
  • Young aspen height can both affect and be affected by browsing: elk browse at a preferred height, so a negative browsing–height correlation is not unambiguous evidence of a trophic cascade.
  • Figure 1 reports a positive relationship between young aspen height and leader length, suggesting height partly reflects growth rate and site productivity; the analysis used 14,095 leader stems from 113 stands in northern Yellowstone, 2007–2017, excluding 40 outliers with leader length above 100 cm.

Methods

  • The paper is a reply to a comment on Brice et al. (2022), synthesizing prior sampling arguments and presenting a new analysis of young aspen height and leader length.
  • The Figure 1 analysis used a generalized additive mixed model with a gamma distribution and log link, modeling height as a cubic function of leader length and browse status, with crossed random intercepts for stand and year.
  • Results are population-averaged fitted values with 95% confidence intervals; the sample came from the same data described by Brice et al. (2022).

Limitations

  • The paper states that 5T sampling is biased and does not represent the broader aspen population response, though it can efficiently detect new aspen recruitment.
  • The paper states that a negative browsing–height correlation is not reliable evidence of a trophic cascade because the causal direction is not unambiguous.
  • Evident from the text: this is a short reply/comment rather than a full primary study, and the Figure 1 analysis is observational and correlational, using an existing sample, so it cannot by itself establish causal cascade strength.

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

Integrity signals

Registry

Crossref · 10.1111/ele.14344

Cites

5 of 5 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

0 of 8 cited sources checked · 0 supported · 0 flagged

  • Findingcitation

    The five-tallest method overestimated overstory aspen regeneration by a factor of four to seven relative to random sampling.

    “Our study quantified the extent of this bias, revealing, for example, that 5T sampling overestimated regeneration of overstory aspen by a factor of 4-7 compared to random sampling (Brice et al., 2022: figure 5).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Brice et al., 2022

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

  • Findingcitation

    The five-tallest sampling method overestimates how much the aspen population responds to wolf reintroduction.

    “Although sampling the five tallest young aspen in a stand is useful for detecting the occurrence of any aspen recruitment, this technique overestimates the population response of aspen to wolf reintroduction.”
    • Also stated in: § Abstract · plain
      “We concluded that 5T sampling overestimated the aspen population response to wolf reintroduction, confirming previous concerns about its limitations.”
    • Also stated in: § Abstract · plain
      “We emphasize that our study characterized the 5T method as biased because it exaggerated estimates of population-level changes in browsing and height of young aspen compared to random sampling.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Brice et al., 2022

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

  • Findingown data

    In randomly sampled young aspen in northern Yellowstone from 2007 to 2017, height is positively related to leader length.

    “Positive relationship between the height and leader length (an index of growth rate and site productivity) of randomly sampled young aspen in northern Yellowstone National Park, 2007-2017.”
    • Also stated in: § Abstract · plain
      “Furthermore, Painter et al.'s claim that height of young aspen is unrelated to leader length (an index of growth rate and site productivity) is contradicted by their data (Painter et al., 2015: appendix A, table A1) and our own (Figure 1).”
    Quote in the paperSubstantial claim
  • Findingreasoning

    Random sampling indicated a weaker trophic cascade than non-random sampling did.

    “Our original conclusion that random sampling described a trophic cascade that was weaker than the one described by non-random sampling is unchanged.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Brice et al., 2022

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

  • Premisecitation

    The five-tallest sampling method can detect the presence of new aspen recruitment.

    “Although sampling the five tallest young aspen in a stand is useful for detecting the occurrence of any aspen recruitment, this technique overestimates the population response of aspen to wolf reintroduction.”
    • Also stated in: § Abstract · plain
      “In summary, we agree with Painter et al. that 5T sampling can efficiently detect the occurrence of new aspen recruitment.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Painter et al., 2023

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

  • May combine two claims. 2
  • Classified as probably not a substantial claim. 2

  • May combine two claims.

    Findingcitation

    Young aspen height is both a cause and a consequence of reduced browsing.

    “We found that height of young aspen is both a cause and an effect of reduced browsing.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Brice et al., 2022

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

  • May combine two claims.

    Interpretationcitation

    Elk browse aspen at a preferred height, and browsing pressure falls as aspen height increases beyond that level.

    “It is a cause of reduced browsing because elk consume aspen at a 'preferred browsing height' beyond which browsing pressure decreases as height increases (Brice et al., 2022: figure 4a).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Brice et al., 2022

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

  • Classified as probably not a substantial claim.

    Interpretationreasoning

    Measuring the strength of the cascade requires a random sampling design that gives a representative estimate of the aspen population response to wolf-caused reductions in elk browsing.

    “Knowledge about the strength of the cascade is also vital, and this requires a random sampling design that provides a representative estimate of the aspen population response to wolfcaused reductions in elk browsing pressure.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: Random sampling provides a representative estimate of the aspen population response.

  • Probably not a substantial claim: may not pass S2 (checkable).

    Interpretationreasoning

    Knowledge about the strength of the cascade is vital.

    “Knowledge about the strength of the cascade is also vital, and this requires a random sampling design that provides a representative estimate of the aspen population response to wolfcaused reductions in elk browsing pressure.”
    Quote in the paperNot a substantial claim
4 of 4

  • The authors state this as an assumption.

    Leader length is an index of growth rate and site productivity.

    “leader length (an index of growth rate and site productivity)”
    “height of young aspen is unrelated to leader length (an index of growth rate and site productivity)”

  • The paper relies on this without support.

    Random sampling provides a representative estimate of the aspen population response.

    “this requires a random sampling design that provides a representative estimate of the aspen population response to wolfcaused reductions in elk browsing pressure.”
2 of 2

Cites

Flagged first, then by citations

Cited by

Sorted by citations

Related

Authors
MacNulty, D. R., Brice, E. M., & Larsen, E. J.
Affiliations
  • Utah State University (US)
  • Cornell University (US)
  • University of Wisconsin–Stevens Point (US)
Corresponding author
MacNulty, Daniel R.
ORCID
Venue
Ecology Letters
Publisher
Wiley
Published
2023-11-27
Volume
27
Issue
1
Pages
e14344-e14344
ISSN
  • 1461-023X (ISSN-L)
  • 1461-0248
DOI
10.1111/ele.14344
OpenAlex
W4389069187
Cited by
6
References
5
Citations by year
2026: 1 · 2025: 4 · 2024: 1
FWCI
0.97
Type
journal article
Language
en
Open access
Licence
onlinelibrary.wiley.com/termsAndConditions#vor (vor)
Locations
  • Ecology Letters (publishedVersion)
  • PubMed (publishedVersion)
  • Digital Commons - USU (Utah State University) (submittedVersion)
Funders
National Science Foundation (DGE‐1633756)
Topics
  • Physical Sciences › Environmental Science › Ecology › Wildlife Ecology and Conservation
  • Physical Sciences › Environmental Science › Ecology › Animal Ecology and Behavior Studies
  • Physical Sciences › Environmental Science › Nature and Landscape Conservation › Ecology and Vegetation Dynamics Studies
Keywords
trophic cascades, wolf reintroduction, non-probability sampling, sampling bias, population response, predator-prey interactions
MeSH terms
Animals, Deer, Predatory Behavior, Wolves, Food Chain

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.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-09-29.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-09-29.

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