Aim: Species with temperature-dependent sex determination (TSD) are particularly vulnerable to climate change because a resultant skew in population sex ratio can have severe demographic consequences and increase vulnerability to local extinction. The Australian central bearded dragon (Pogona vitticeps) has a thermosensitive ZZ male/ZW female system of genetic sex determination (GSD). High incubation temperatures cause reversal of the ZZ genotype to a viable female phenotype. Nest temperatures in the wild are predicted to vary on a scale likely to produce heterogeneity in the occurrence of sex reversal, and so we predict that sex reversal will correlate positively with inferred incubation conditions.

Location: Mainland Australia

Methods: Wild-caught specimens of P. vitticeps vouchered in museum collections and collected during targeted field trips were genotypically and phenotypically sexed to determine the distribution of sex reversal across the species range. To determine whether environmental conditions or genetic structure can explain this distribution, we infer the incubation conditions experienced by each individual and apply a multi-model inference approach to determine which conditions associate with sex reversal. Further, we conduct reduced representation sequencing on a subset of specimens to characterise the population structure of this broadly distributed species.

Results: Here we show that sex reversal in this widespread Australian dragon lizard is spatially restricted to the eastern part of the species range. Neither climatic variables during the inferred incubation period nor geographic population genetic structure explain this disjunct distribution of sex reversal. The main source of genetic variation arose from isolation by distance across the species range.

Main Conclusions: We propose that local genetic adaptation in the temperature threshold for sex reversal can counteract the sex-reversing influence of high incubation temperatures in P. vitticeps. Our study demonstrates that complex evolutionary processes need to be incorporated into modelling biological responses to future climate scenarios.

README for genetic and incubation condition analysis in Diversity and Distributions manuscript "Evolving thermal thresholds explain the distribution of temperature sex reversal in an Australian dragon lizard:"

SPECIMEN COLLECTION INFORMATION AND ANALYSIS

Field_Pogona_Data.xlsx
Contains collection information and collated environmental data (from the Scientific Information for Landowners dataset https://www.longpaddock.qld.gov.au/silo/) for the inferred incubation period for each individual. The first sheet in the excel workbook describes the data contained within each column, and this data is used in the "Inferred_Incubation_Conditions.R" script.

pogona_dart_genlight
RDS object containing the raw DArT-seq data for each of the specimens sequenced for population genetic analyses. This object is read into the R script "pogona_dart_analysis_Aug2020.R" for all subsequent analyses, mainly using the R package "dartR".

pogona_dart_analysis_Aug2020.R
R script containing the analyses used to investigate the population genetics of Pogona vitticeps. Based mostly on the functions within the R package dartR, uses the "pogona_dart_genlight" object.

Inferred_Incubation_Conditions.R
R script containing analyses used to investigate the association between sex reversal and inferred incubation temperatures. Uses the data in "Field_Pogona_Data.xlsx"

silo_data_extraction.py
Python code for the extraction of climatic data (SILO climate dataset) from relevant dates for each collected specimen. Uses data from "Field_Pogona_Data.xlsx" and relevant SILO climate netCDF files

CONSTANT TEMPERATURE EQUIVALENT ANALYSIS

Please note: detailed instructions on how to run each piece of code can be found within each piece of code's comments (primarily at the top and in the input parameters sections). This is an overview of the pipeline rather than a replication of those instructions.

To go from climate data to CTE:
1: Climate data: Download elevation and SILO climate data (variables listed in the Castelli_MicroclimateAndCTE_RandomLocations.R file). This has been completed and climate data are provided in csv files with the prefix "randomlocations_". This includes temperature, rainfall, relative humidity, radiation and vapour pressure.

2: Wind data: Download the near speed wind data and format it as detailed in Castelli_NearSurfaceWindData_extraction.R. Run that file to extract the wind speed data for the locations and times of interest.

3: SLGA grids: Download the soil grids (.tiff) as documented in Castelli_MicroclimateAndCTE_RandomLocations.R

4: To go from climate data to soil temperature data (calculating the microclimate using NicheMapR): Open Castelli_MicroclimateAndCTE_RandomLocations.R. Follow the instructions within and change the input parameters to match the file paths on your computer. Run or source the file.

5: To go from soil temperature data to CTE: Open Castelli_SomaCTE_RandomLocations.R. Follow the instructions to update the input parameters to match your computer. Run or source this file to calculate the mean CTE per year (and the daily CTE if required). This file can be run sequentially (very slow), or in parallel (calculating multiple sites at once, much faster). It took about 5 days to process 1000 sites using 20 cores.

6: Figure1_SupplementaryData.xlsx
Contains the constant temperature equivalent (CTE) estimates for each randomly selected location within the range of Pogona vitticeps (displayed in Figure 1 of the manuscript). The first sheet in the excel workbook explains the data contained within each column. This file contains the summarised output of the steps outlined above.