Increasing species richness along elevational gradients is associated with niche packing in bat assemblages

1. The change in species richness along elevational gradients is a well-known pattern in nature. Niche theory predicts that increasing species richness in assemblages can either lead to denser packing of niche space (‘niche packing’) or an expansion into its novel regions (‘niche expansion’). Traditionally, these scenarios have been studied using functional traits, but stable isotopes provide advantages such as identifying the degree of resource specialisation or niche partitioning among functionally similar species.

2. In this study, we evaluate the relevance of niche packing vs. niche expansion by investigating stable carbon and nitrogen isotopic niche width and overlap among 23 bat species from six functional groups across a 1500 m elevational gradient in the Himalaya.

3. Our results suggest that an increase in species richness in the low elevation is accompanied by small niche width with high overlap, whereas the high elevation assemblage shows large niche width with low overlap among functional group members. At the functional group level, edge-space foraging, trawling, and active gleaning bats have the highest niche width while passive-gleaning bats that are only found in high elevations are isotopic specialists showing low overlap with other groups. Edge and open-space foraging bats showed idiosyncratic changes in niche width across elevations. We also find that the niches of rhinolophid bats overlap with edge-space and open-space foraging bats despite their unique functional traits.

4. These results support the idea that, at low elevations, high species richness is associated with niche packing while at high elevations, strong niche partitioning prevails in dynamic and resource-poor environments. We conclude that although high-elevation animal assemblages are often ‘functionally underdispersed’, i.e. show homogenous functional traits, our approach based on stable isotopes demonstrates niche partitioning among such functionally similar species.

Methods
Please refer to the manuscript for detailed methods on how the data were collected. The isotope data on bats, plants, and insects have not been processed. The other two datasets (Layman metrics and SEAb) were calculated from the isotope data using the SIBER R package.