Interpretable and predictive models to harness the life science data revolution DOI Creative Commons
Joshua P. Jahner, C. Alex Buerkle, Dustin Gannon

et al.

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2024, Volume and Issue: unknown

Published: March 17, 2024

Abstract The proliferation of high-dimensional data in ecology and evolutionary biology raise the promise statistical machine learning models that are highly predictive interpretable. However, commonly burdened with an inherent trade-off: in-sample prediction outcomes will improve as additional predictors included model, but this may come at cost poor accuracy limited generalizability for future or unsampled observations (out-of-sample prediction). To confront problem overfitting, sparse can focus on key by correctly placing low weight unimportant variables. We competed nine methods to quantify their performance variable selection using simulated different sample sizes, numbers predictors, strengths effects. Overfitting was typical many simulation scenarios. Despite this, out-of-sample converged true target simulations more observations, larger causal effects, fewer predictors. Accurate support process-based understanding be unattainable realistic sampling schemes evolution. use our analyses characterize attributes which is possible, illustrate how some achieve while mitigating extent overfitting.

Language: Английский

Aligning renewable energy expansion with climate-driven range shifts DOI
Uzma Ashraf, Toni Lyn Morelli, Adam B. Smith

et al.

Nature Climate Change, Journal Year: 2024, Volume and Issue: 14(3), P. 242 - 246

Published: March 1, 2024

Language: Английский

Citations

11
Mapping climatically suitable areas for tree species across the globe: comparing TreeGOER-based maps with previous maps generated for nine selected eucalypt species DOI Creative Commons
Trevor H. Booth,

Tom Jovanovic

Australian Forestry, Journal Year: 2025, Volume and Issue: unknown, P. 1 - 9

Published: March 19, 2025

Language: Английский

Citations

0
Biotic interactions shape the realised niche of toxic cyanobacteria DOI Creative Commons

Pinelopi Ntetsika,

Stefanie Merkli, Ewa Merz

et al.

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2025, Volume and Issue: unknown

Published: April 20, 2025

Abstract Cyanobacterial blooms increasingly threaten vital freshwater ecosystems, with harmful impacts exacerbated by climate change and eutrophication. Despite extensive research on temperature nutrient effects, our predictive capacity remains limited. We propose that this limitation stems from insufficient understanding of how biotic interactions modify cyanobacterial responses to abiotic conditions. Using five years daily monitoring data a eutrophic lake state-space reconstruction modelling, we show co-occurring plankton species fundamentally reshape the realised niche bloom-forming cyanobacteria. Biotic shift thresholds up 13°C phosphorus requirements over 20 μg/L—effects substantial enough determine whether environmental conditions support or prevent in Microcystis Dolichospermum . Grazing inhibits bloom formation across taxa, while facilitation other phytoplankton may allow at unexpectedly low temperatures phosphate concentrations. These findings address fundamental gap—how shape niches natural systems—while offering practical insights for management. By integrating into programs models, can improve forecasting accuracy develop targeted interventions complement traditional control approaches. parallel recent advances ecology suggesting role mediating species’ change.

Language: Английский

Citations

0
Future Climate Shifts for Vegetation on Australia's Coastal Islands DOI Creative Commons
David Coleman, Mark Westoby, Julian Schrader

et al.

Global Change Biology, Journal Year: 2025, Volume and Issue: 31(5)

Published: May 1, 2025

ABSTRACT Small coastal islands serve as replicated units of space that are useful for studying community assembly. Using a unique database holding information on comprehensive vegetation surveys > 840 small fringing the whole continent Australia, we investigated extent to which conditions will change plants Australia's over next 80 years in terms their temperature envelopes and inferred changes vapour pressure deficit (VPD). We found ~40% island plant populations experience mean annual temperatures beyond current envelope. However, defined by VPD extreme monthly unlikely be exceeded, highlighting islands' potential act climate refugia. Large species with slow life histories poor dispersal traits were most likely warmer temperatures, although this proved driven correlations these latitude (closer equator) smaller range sizes. no evidence warm edge extinction or poleward migration across response 0.5° warming since year 2000. These results have applications monitoring conservation efforts under fragmented habitats everywhere.

Language: Английский

Citations

0
Predicting current and future potential distribution of Changnienia amoena in China under global climate change DOI Creative Commons
Xingjian Liu,

Qimeng Sun,

Tingting Li

et al.

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: May 21, 2025

Language: Английский

Citations

0
Assessing the Impact of Climate Change on Habitat Shifts of Typical Endemic Relict Trees in China DOI Creative Commons
Ming Li, Junbang Wang, Xiujuan Zhang

et al.

Global Ecology and Conservation, Journal Year: 2025, Volume and Issue: unknown, P. e03643 - e03643

Published: May 1, 2025

Language: Английский

Citations

0
Climate suitability for the moisture-sensitive conifer species may not be universally declining in a warming world DOI
Bo Wang, Tuo Chen, Guobao Xu

et al.

Agricultural and Forest Meteorology, Journal Year: 2024, Volume and Issue: 361, P. 110328 - 110328

Published: Nov. 28, 2024

Language: Английский

Citations

1
Biodiversity forecasting in natural plankton communities reveals temperature and biotic interactions as key predictors DOI Creative Commons
Ewa Merz, Francesco Pomati, Serguei Saavedra

et al.

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2024, Volume and Issue: unknown

Published: Oct. 13, 2024

Summary As natural ecosystems experience unprecedented human-made degradation, it is urgent to deliver quantitative anticipatory forecasts of biodiversity change and identify relevant biotic abiotic predictors. Forecasting has been challenging due their complexity, chaotic nonlinear nature the availability adequate data. Here, we use four years daily abundance a complex lake planktonic ecosystem its environment model forecast metrics. Using state-of-the-art equation-free modelling technique, community richness turnover with proficiency greater than constant predictor several generations ahead (30 days). Short-term improve substantially using predictors (i.e., autoregressive term or richness). Long-term require more set variables interactions), depends strongly on including such as water temperature. Depending horizon, can interact nonlinearly synergistically, enhancing each other’s effects Our findings showcase challenges forecasting in stress importance monitoring focal anticipate undesired changes.

Language: Английский

Citations

0
Author Correction: Aligning renewable energy expansion with climate-driven range shifts DOI Creative Commons
Uzma Ashraf, Toni Lyn Morelli, Adam B. Smith

et al.

Nature Climate Change, Journal Year: 2024, Volume and Issue: unknown

Published: Dec. 5, 2024

Language: Английский

Citations

0
Interpretable and predictive models to harness the life science data revolution DOI Creative Commons
Joshua P. Jahner, C. Alex Buerkle, Dustin Gannon

et al.

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2024, Volume and Issue: unknown

Published: March 17, 2024

Abstract The proliferation of high-dimensional data in ecology and evolutionary biology raise the promise statistical machine learning models that are highly predictive interpretable. However, commonly burdened with an inherent trade-off: in-sample prediction outcomes will improve as additional predictors included model, but this may come at cost poor accuracy limited generalizability for future or unsampled observations (out-of-sample prediction). To confront problem overfitting, sparse can focus on key by correctly placing low weight unimportant variables. We competed nine methods to quantify their performance variable selection using simulated different sample sizes, numbers predictors, strengths effects. Overfitting was typical many simulation scenarios. Despite this, out-of-sample converged true target simulations more observations, larger causal effects, fewer predictors. Accurate support process-based understanding be unattainable realistic sampling schemes evolution. use our analyses characterize attributes which is possible, illustrate how some achieve while mitigating extent overfitting.

Language: Английский

Citations

0