Multi-phase modelling of potential drivers for the range expansion of the golden jackal (Canis aureus).

Ecological niche models are often snapshots in space and time, providing little information on how niches change in response to changing environmental conditions and species interactions over time. The golden jackal has shown one of Europe's most drastic expansions in recent decades. However, modelling the drivers behind the expansion has often been limited to a single time period and focusing mainly on land cover and current climatic characteristics and its changes rather than incorporating biotic interactions. Therefore, we aim to identify the drivers that influence the species distribution and assess their impacts on each phase of its expansion process, with a focus on competitor species and consumed resources. We applied an ecological niche modelling approach to investigate the species-environment relationship including biotic interactions across each phase of the expansion process, utilising regression and MaxEnt models with hunting bag data as occurrence records of the species in 10 × 10 km grid cells. We fitted presence-absence, abundance, relative abundance, density, and presence-only models for each five-year period from 1997 to 2021 in Hungary. As the expansion progressed, more predictors became significant, indicating increasing complexity in newly colonised areas. Red deer harvest and elevation consistently emerged as the most significant and influential predictors throughout the jackal's expansion across all models. While red deer are not direct prey for jackals, the leftover viscera from hunting confirms existing evidence that jackals heavily rely on anthropogenic food resources. Additionally, the negative relationship with elevation highlights the golden jackals' selection for lower-altitude habitats. Our results show that modelling during the early phases of an invasion may fail to capture the full spectrum of environmental conditions shaping a species' distribution. This emphasises the importance of accounting for distinct phases of the invasion process and incorporating biotic interactions to improve the accuracy of niche and species distribution models.