How Individual Variation Shapes Ecological Niches in Two Pipistrellus Bat Species

Niche partitioning is a key mechanism explaining species coexistence and biodiversity. While different axes of niche partitioning can account for species coexistence, variation between and within individuals can significantly influence this process. This empirical study investigates niche partitioning in a comprehensive framework using two co-occurring bat species as model organisms: Nathusius’ pipistrelle Pipistrellus nathusii and Kuhl’s pipistrelle P. kuhlii. Pipistrellus kuhlii is a recent invader of the study area, rapidly expanding its geographic range. As human-induced alterations in land cover and climate are likely to accelerate this range expansion, understanding the mechanisms enabling the co-occurrence of these species is crucial. We examined niche partitioning across habitat, time, and resource use, considering variation between species, between-individual, and within-individual levels. Specifically, we analysed the movement patterns of 58 individuals of both species and employed metabarcoding on guano samples to assess resource use. Our results demonstrate that individual variation exceeded species-level differences and varied between species.Pipistrellus nathusii exhibited greater between-individual variation, while the expanding P. kuhlii displayed stronger within-individual variation, which may contribute to its range expansion. Our study suggests an important role of within-individual variation in the ongoing expansion of an invading bat species, reshaping animal communities in the context of global change.

Data and scripts of “The ecological niche and individual variation of a rapidly expanding bat”

a_TriangulationEstimation

This folder contains the R-scripts required to triangulate the pre-filtered data. First step is to pre-filter data based on the interval between consecutive signals. Then, the bearings are calculated using a cosine approach. Finally, the bearings are triangulated using Azimuthal Telemetry Models (ATM).

The fits of the ATM have a size of 56.42 GB. Thus, they are not provided on Dryad, but can be shared on request.

Input: Raw data

output: Triangulated location estimates including inaccuracy estimates

b_filterRawPoints

This folder contains the scripts to filter the triangulated locations based on activity (moving vs roosting) and speed.

Input: Triangulated location estimates including inaccuracy estimates

Output: Filtered location estimates

c_MovementModels

This folder contains the scripts to transform the location data to dynamic Brownian Bridge movement models and burst the data to 30 min, 10 min, and 1 hour intervals.

Input: Filtered location estimates

Output: Movement models, burst to 30 min, 10 min, and 1 hour intervals

d_landscapePreparation

This folder contains the scripts to crop the land cover data and calculate selection ratios of the respective land cover types.

Input: land cover data (Corine, Land Burgenland), movement models

Output: Temporal explicit selection ratios

e_ nicheModels

This folder contains the scripts to build niche models based on the temporal explicit selection ratios

Input: Temporal explicit selection ratios

Output: Individual niche models

f_metabarcoding

This folder contains the analysis of the metabarcoding.

Input: Output of metabarcoding analysis

Output: PERMANOVA of metabarcoding results, NMDS and Pianka’s niche overlap