EUVIMEDEUVIMED entdecken
Crossref

Integrating Phytosociological Proxies to Improve Predictions of Tree Species’ Local Ecological Performance: Potentials and Limitations

João Paulo Czarnecki de Liz, Nelson Thiffault, Loïc D'Orangeville, Nicholas C. Coops, Alexis Achim

Canadian Journal of Forest Research · 21.11.2026

DOI: 10.1139/cjfr-2026-0221 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Canadian Journal of Forest Research, online erschienen am 21.11.2026. Verfasst von João Paulo Czarnecki de Liz und 4 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Intensifying disturbances and climate change are disrupting forest ecosystems globally, making science-based adaptation strategies essential in forest management. Species distribution models (SDMs) support this task, but their predictive accuracy is often limited by a primary reliance on abiotic variables. We evaluated if the incorporation of tree species co-occurrence data, a proxy for biotic interactions and habitat filtering, can improve the prediction of species’ local ecological performance (realized abundance and dominance, measured as Importance Value, IV) in the boreal-temperate ecotone of Quebec, Canada. Using Random Forest regression, we compared abiotic-only versus abiotic + co-occurrence models for seven ecologically important tree species. Co-occurrence predictors consistently improved generalization to independent test data (test R² from 0.01–0.44 to 0.07–0.73), reduced prediction bias (from -1.23 to -0.88 for Acer saccharum; from -1.20 to -0.22 for Picea mariana), and frequently displaced broad climatic predictors in importance rankings. These findings suggest that using the IV of the most important species of the community can help capture fine-scale, deterministic filters that influence habitat suitability. However, unexplained variance remains, highlighting the inherent stochasticity and uncertainty involved in predicting species dominance/abundance at the stand-level. Despite persistent uncertainties, forest-inventory co-occurrence data offer a scalable baseline for habitat suitability predictions, supporting climate-adaptive forestry.

Zitieren

APA

Liz, J. P. C. D., Thiffault, N., D'Orangeville, L., Coops, N. C., & Achim, A. (2026). Integrating Phytosociological Proxies to Improve Predictions of Tree Species’ Local Ecological Performance: Potentials and Limitations. Canadian Journal of Forest Research. https://doi.org/10.1139/cjfr-2026-0221

Vancouver

Liz JPCD, Thiffault N, D'Orangeville L, Coops NC, Achim A. Integrating Phytosociological Proxies to Improve Predictions of Tree Species’ Local Ecological Performance: Potentials and Limitations. Canadian Journal of Forest Research. 2026. doi:10.1139/cjfr-2026-0221

BibTeX
@article{Liz2026,
  author = {Liz, João Paulo Czarnecki de and Thiffault, Nelson and D'Orangeville, Loïc and Coops, Nicholas C. and Achim, Alexis},
  title = {Integrating Phytosociological Proxies to Improve Predictions of Tree Species’ Local Ecological Performance: Potentials and Limitations},
  journal = {Canadian Journal of Forest Research},
  year = {2026},
  doi = {10.1139/cjfr-2026-0221}
}
Kostenlos starten →