Data from: Demographic performance of Cerrado lizards: A test of the center-periphery hypothesis
Abstract
The center-periphery hypothesis (CPH) states that demographic performance is highest in the center of species’ ranges and decreases as the distance from the center increases due to declining environmental suitability. We tested the predictions under the CPH for two lizard lineages (Ameiva ameiva, Tropidurus itambere, and T. madeiramamore) with distinct distribution patterns. We assessed demographic performance using body condition and parasite information as indicators of demographic performance. The body condition refers to Peig and Green’s Scaled Mass Index, which associates body length and mass to indicate body status. Samples were collected from core Cerrado localities (South American savannas) and peripheral isolates in southwestern Amazonia. We built generalized linear mixed models with demographic performance as the response variable, environmental (climate, elevation, and soil variables) and spatial variables (landscape parameters and distance to the Cerrado’s center and periphery, e.g., centrality and peripherality), lizard genus and their interactions as fixed effects, and municipality and locality as random effects. We used Generalized Dissimilarity Modeling and variance partitioning to check the importance of geographic distance, environmental and spatial variables, and the dissimilarity among lizard communities on parasite beta diversity.
(https://doi.org/10.5061/dryad.g79cnp60t)
The center-periphery hypothesis (CPH) states that demographic performance is highest in the center of species’ ranges and decreases as the distance from the center increases due to declining environmental suitability. We tested the CPH predictions for two lizard lineages (Ameiva ameiva, Tropidurus itambere, and T. madeiramamore) with distinct distribution patterns. We assessed demographic performance using body condition and gastrointestinal parasite information. The body condition refers to Peig and Green’s Scaled Mass Index, which associates body length and mass as a way to indicate body status. Our samples are from localities at the core Cerrado (South American savannas) and peripheral isolates in southwestern Amazonia. We built generalized linear mixed models with demographic performance as the response variable, environmental (climate, elevation, and soil variables) and spatial variables (landscape parameters and distance to the Cerrado’s center and periphery, e.g., centrality and peripherality), lizard genus and their interactions as fixed effects, and municipality and locality as random effects. We used Generalized Dissimilarity Modeling and variance partitioning to check the importance of geographic distance, environmental and spatial variables, and the dissimilarity among lizard communities on parasite beta diversity. Results showed lineage-specific responses. Ameiva ameiva exhibited stable performance across sites, consistent with expectations under the CPH for central populations. Tropidurus exhibited better performance in isolates, suggesting ecological release, contrary to predictions under the CPH. Finally, parasite beta diversity was driven mainly by soil variation and centrality, indicating the importance of environmental and spatially structured variation and little to no importance of biotic or climatic variation.
Description of the data and file structure
These data refer to the lizard gastrointestinal parasite community, morphometric data, and the lizard community. These were sampled in isolates from the core Cerrado at Goiás and the peripheral isolates at Rondônia. We provide a Quarto document with the code for analysis in the R (and R Studio) environment. Unavailable data are left as empty cells.
All files are provided in a compressed folder: Files.zip. After extraction, the folder contains the following subdirectories and files:
Data folder - contains the data necessary for running the analyses.
- DataParacatu&Rondonia/ - Data from sampled localities.
-
lizard_morphometry_final.xlsx: Morphometric data for sampled lizards.
GRCOLLI- Collection ID (unique identifier for specimen/record).SVL- Snout-vent length (mm).TAIL- Tail length (mm).WT- Body weight (g).HL- Head length (mm).HW- Head width (mm).HD- Head depth (mm).FLL- Forelimb Length (mm).HLL- Hindlimb Length (mm).LOCALITY- Municipality name (official IBGE designation, in Portuguese).patch- Patch ID.STATE- State (federative unit), IBGE designation, in Portuguese.spGenus- Genus.spName- Specific epithet.NAs refer to non-available data.
-
LizardsParacatuRondonia: lizard sampling.
patch- Patch ID.localityNames- Municipality name (official IBGE designation, in Portuguese).Remaining columns - Species counts as number of individuals. Columns are labeled by concatenated genus + species (e.g., AmeivaAmeiva)
-
patchData2.xlsx: landscape variables. See the manuscript for further descriptions of the measured variables.
patch- Patch IDlat_dms- Latitude (degrees, minutes, seconds)long_dms- Longitude (degrees, minutes, seconds)areaHa- Fragment area (hectares)nPatch- Number of internal habitat patchespatchDens- Inner fragmentation (see manuscript for calculation method)patchBuffer- Fragment connectivityperimeter- Fragment perimeter (km)edge- Edge length (km)
- ParasiteData/ -
-
analysisData.xls: Gastrointestinal parasite sampling data.
We used data from the first sheet
[Species]Genus- Genus.SpeciesName- Specific epithet.Local- Patch ID.Locality Name- Municipality name (official IBGE designation, in Portuguese).Remaining columns - Parasite count.
- RDSfiles/ - Derived and organized environmental and biological variables necessary for running the analyses.
- lizard_by_patch.RData: lizard sampling by locality.
- parasite_data_clean.RData: parasite sampling by locality.
- patch_data.RData: landscape variables by locality.
- soil_by_patch_df.RData: soil variables by locality.
- ameiva_records.csv: Species records of Ameiva ameiva used for mapping.
- tropidurus_spp_records.csv: Species records of Tropidurus spp. Used for mapping.
ameiva_records.csv and tropidurus_records.csv share the same column structure:
Genera - Genus.
Scientific Name - Full binomial species name.
Latitude - Latitude of occurrence record (in decimal degrees).
Longitude - Longitude of occurrence record (in decimal degrees).
State - State (federative unit, IBGE designation, in Portuguese).
Municipality - Municipality name (IBGE designation, in Portuguese).
NAs refer to non-available data.
- Code_for_manuscript.qmd – Quarto document with annotated R code to reproduce all analyses. Includes GLMMs for body condition and parasitism, and GDM for parasite beta diversity.
Sharing/Access information
Original environmental datasets were obtained from external repositories and are not redistributed here. Only derived values (site-level extractions and PCA scores) are included. Users should download the raw data directly from the sources below:
- IBGE cartographic data: http://geoftp.ibge.gov.br/
- WorldClim bioclimatic variables: https://www.worldclim.org/
- SoilGrids soil variables: https://soilgrids.org/
References
- Instituto Brasileiro de Geografia e Estatística (IBGE). 2019. Biomes of Brazil: shapefiles at 1:250,000 scale. Available at: http://geoftp.ibge.gov.br/
- Fick, S. E. and Hijmans, R. J. 2017. WorldClim 2: New 1‐km spatial resolution climate surfaces for global land areas. – Int. J. of Climatol. 37: 4302-4315.
- Poggio, L., de Sousa, L. M., Batjes, N. H., Heuvelink, G. B. M., Kempen, B., Ribeiro, E., Rossiter, D. 2021. SoilGrids 2.0: Producing soil information for the globe with quantified spatial uncertainty. – SOIL 7: 217-240.
This data refers to the lizard gastrointestinal parasite community, morphometric data, and the lizard community. These were sampled in isolates from the core Cerrado at Goiás and peripheral isolates at Rondônia. We used body condition and gastrointestinal parasitism as surrogates of lizard demographic performance.
To assess the predictions of the Center-Periphery Hypothesis, we built generalized linear mixed models with demographic performance as the response variable, environmental (climate, elevation, and soil variables) and spatial variables (landscape parameters and distance to the Cerrado’s center and periphery, e.g., centrality and peripherality), lizard genus and their interactions as fixed effects, and municipality and locality as random effects.
To get a more nuanced understanding of the variation in parasitism, we used Generalized Dissimilarity Modeling and variance partitioning to check the importance of geographic distance, environmental and spatial variables, and the dissimilarity among lizard communities on parasite beta diversity.