/urbanBias

study of urban Bias in CS data

Primary LanguageRMIT LicenseMIT

urbanBias

A study of the strength of urban Bias in CS data and implications of changes in this bias for population trend estimation.

Simulations

In this folder are the main scripts:

  • spatialBias.R - the main script to run each scenario under the standard settings (urban avoiding species, perfect detection)
  • spatialBias_speciesEffects.R - as the main script except to generate the patterns for species either neutral or responding positively to urban cover.
  • spatialBias_function.R - the functions to generate datasets and analyse. This script is called within the above scripts.

Models

These are the text files for the JAGS occupancy models. These are not used in the urban bias paper.

Methods

This folder just contains some bits of code for different methods.

Empirical-data

This pulls together the data from the real-world datasets to analyse the spatial bias and draw the empirical data plot. The first analysis was done within the empirical_bias_Plots.R The analysis was extended using more datasets and using spAMM - which meant moving to the HPC Hence all the other files are script to test

  • the effect of urban cover and pa area on site visitation in each year (HPC_testPABias, HPC_testUrbanBias)
  • test the main effect of urban cover across all years (HPC_testmainPABias, HPC_testmainUrbanBias)
  • test the interaction between the effect of urban cover and year (HPC_testinteractionUrbanBias, HPC_testinteractionPABias) HPC_models just pulls in the models from the HPC. recordTS.R plots a time series data_processing contains script to process each dataset that was downloaded.