# Global Patterns of Niche Changes in Alien Mammals: Potential Drivers and Significance for Invasion Projections

## TL;DR

Alien mammals mostly stick to their native climate niches, but often don't fully colonize new regions, affecting how well models predict their spread.

## Contribution

The study quantifies niche changes in alien mammals and identifies drivers affecting invasion projections using a robust comparative framework.

## Key findings

- Niche expansion is rare and modest, while niche unfilling is common and pronounced.
- Niche unfilling indicates large pools of suitable yet unoccupied climatic conditions in new regions.
- SDM transferability declines with niche expansion and unfilling, affecting invasion prediction accuracy.

## Abstract

Biological invasions are a major driver of global change, and prevention is the most effective mitigation strategy. Bioclimatic species distribution models (SDMs) are widely used to estimate invasion risk, assuming that species retain their realized native climatic niches after introduction. We tested this assumption for 194 alien mammal species established across 11 zoogeographic realms, examining realized niche changes, their drivers, and significance for invasion projections. We used a robust ordination framework to compare native and alien niches in 337 species‐by‐realm niche comparisons and quantify niche expansion, the proportion of the alien niche not overlapping with the native niche, and niche unfilling, the proportion of the native niche not overlapping with the alien niche. We then applied Generalized Linear Mixed Models (GLMMs) with multi‐model inference to test how species attributes, invasion history, and environmental context are associated with expansion and unfilling. Additionally, we evaluated the transferability of SDMs built on native presences to receiving regions using multiple metrics and used GLMMs to assess how niche changes may affect it. Niche expansion was rare and modest, whereas niche unfilling was common and pronounced. Niche expansion declined with increasing human disturbance, larger native range size, and introductions within similar communities, but increased with higher introduction effort. Niche unfilling decreased with greater introduction effort and longer residence time, but increased with alien insularity, human disturbance, and native range loss. SDM transferability was generally good, but it declined with niche expansion, as alien presences fell outside native‐like suitable areas, and with unfilling, because suitable areas remained unoccupied under colonization lags. Proactive management can rely on SDMs to anticipate future spread and should prioritize species showing high niche unfilling, indicating substantial spread potential, and any evidence of niche expansion, which makes spread harder to anticipate.

Alien mammals largely retain their realized native climatic niche, yet often show incomplete colonization of receiving zoogeographic regions. Across 337 species‐by‐realm niche comparisons, niche expansion is generally low and concentrated in a few zoogeographic realms, whereas niche unfilling is widespread, indicating large pools of suitable yet unoccupied climatic conditions. Generalized linear mixed models mainly associate niche expansion with higher introduction effort, smaller native ranges, and lower community similarity, and niche unfilling with lower introduction effort, shorter residence time, and higher human disturbance. Species distribution models built on native presences generally transfer well to the receiving regions, but niche expansion reduces performance as alien presences occur outside native‐like suitable conditions, while niche unfilling mainly lowers metrics penalized by suitable but unoccupied areas.