Data from: Individual asymmetric competition responses across multidimensional niches may enable coexistence of closely related species
Data files
Jun 04, 2025 version files 615.58 KB
-
allotu.csv
153.42 KB
-
GPS_data.csv
408.51 KB
-
OTU_INFO.xlsx
48.84 KB
-
README.md
4.81 KB
Abstract
Many studies have focused on niche differentiation at the population level to explain the coexistence of similar species. However, information on how individual-level processes across multidimensional niches shape community dynamics and species coexistence, especially for nocturnal, small, and highly mobile animals, is limited.
In this study, we employed a combination of metabarcoding and GPS tracking technologies to investigate the coexistence mechanisms between two sympatric bat species, Hipposideros armiger and H. pratti, by analyzing individual niche responses to competition across temporal, spatial, and dietary dimensions.
Results showed that (1) traditional population-level analysis revealed no significant niche partitioning along single dimensions. However, (2) individual-level analyses uncovered sophisticated coexistence mechanisms through asymmetry responses to interspecific competition via temporal and dietary dimensions within limited habitat ranges. (3) This asymmetry response ensures stability of complementary relationships through coordinated interactions across three dimensions. (4) Intraspecific competition contributes to species asymmetry stabilization by modifying temporal activity patterns of both species, thereby reducing interspecific competition and facilitating coexistence.
In conclusion, H. armiger and H. pratti achieved stable coexistence through coordinated responses across all three niche dimensions, with individuals demonstrating complementary patterns between dietary utilization and temporal activity, rather than single-dimension partitioning. Our work provides a comprehensive framework for understanding how individual-level multidimensional niche adjustments and asymmetric competitive responses facilitate stable coexistence in sympatric species.
Dataset DOI: https://doi.org/10.5061/dryad.j0zpc86r1
Description of the data and file structure
This dataset supported a study on the coexistence strategies of Hipposideros armiger and H. pratti in Xiaonanhai Town, Nanzheng District, Hanzhong City, Shaanxi Province, China, from July to August 2022. This study used GPS logers and fecal DNA-metabarcoding to monitor population and individual niche differentiation across multiple dimensions (temporal, feeding, and spatial) of 17 adult individuals caught in fish holes to investigate their behavioral responses in the face of interspecific and intrspecific competition. The data capture the contribution of individual niche plasticity and competitive asymmetry to species coexistence, It suggested that stable coexistence between H. arabica and H. patelli is maintained through multiple complementary mechanisms, including asymmetric competition stabilized by intraspecific competition and dimensional coordination, strong niche trade-offs, and species-specific competitive strategies. This study provides a new idea and framework for exploring the coexistence strategies of similar species
Files and variables
File: GPS_data.csv
Description:
Variables
- id: A unique alphanumeric identifier assigned to each tracked bat (e.g., D11, H11).
- timestamp: Recording date in YYYY/MM/DD format, covering the study period from June 2022 to August 2022.
- longitude: Geographic longitude coordinate (decimal degrees) of the recorded location.
- latitude: Geographic latitude coordinate (decimal degrees) of the recorded location.
- speed: Instantaneous movement speed (units: km/h) of the individual at the recorded timestamp.
- angle: Movement heading/direction (units: degrees from North) of the individual at the time of recording.
- height: Elevation above ground level (units: meters ) at the recorded position.
- temp: Ambient temperature (units: °C) logged by the tracking device.
- v: Current battery voltage (units: volts) of the tracking device.
File: allotu.csv
Description:
This file contains the OTU (Operational Taxonomic Unit) read count matrix from dietary analysis. The first column lists OTU identifiers (OTU1 through OTU1445), with each OTU's full taxonomic classification available in the accompanying OTU_INFO file. Subsequent columns represent individual bat specimens, where column headers correspond to individual IDs (some of which can be cross-referenced with individuals in the GPS_data file). The numerical values in the matrix indicate the number of sequencing reads assigned to each OTU for each individual, reflecting relative abundance in dietary samples. Missing data (if present) are represented by NA.
Variables
- OTU: OTU identifiers (OTU1 through OTU1445), with each OTU's full taxonomic classification available in the accompanying OTU_INFO file.
- Subsequent columns: individual bat specimens, where column headers correspond to individual IDs (some of which can be cross-referenced with individuals in the GPS_data file). The numerical values in the matrix indicate the number of sequencing reads assigned to each OTU for each individual.
File: OTU_INFO.csv
Description: This file provides the full taxonomic classification for each OTU (Operational Taxonomic Unit) listed in allotu.csv.
Variables
- *OTU：*OTU identifiers (e.g., OTU1, OTU2), which correspond exactly to the OTU numbering in the read count matrix.
- *source：*the taxonomic assignment for each OTU using a standardized eight-rank format: domain (d), kingdom (k), phylum (p), class (c), order (o), family (f), genus (g), and species (s). For example, the entry d__Eukaryota;k__Metazoa;p__Arthropoda;c__Insecta;o__Coleoptera;f__Carabidae;g__Apenetretus;s__Apenetretus_hsueshanensis indicates a beetle species from the family Carabidae.
All analyses were performed in R 4.2.1 (R Core Team, 2022) with the following packages:
adeHabitatHR: Kernel density estimation for home range calculation and overlap analysis (95% UD contours)vegan: Dietary niche metrics (Shannon alpha diversity, Levins' niche width) and ANOSIM for compositional differencesdismo+MuMIn: Logistic regression with AIC-based model selection (ΔAIC < 2 criterion)brms: Bayesian hierarchical models with weakly informative priors (Stan backend)overlap: Temporal activity pattern overlap estimation (Dhat4 estimator)
Supplementary Analyses
Time-series geometry parameters were extracted using MATLAB 2022b (MathWorks) .