MoveApps
Github repository: https://github.com/nilanjanchatterjee/Step_selection_Analysis
The App models observed steps with the randomly generated steps using the same distribution estimated from the observed steps. Currently the app accepts user provided raster files and a fallback file of Distance to road is provided for the Yellowstone to Yukon region with a resolution of 0.01 degree.
The app models steps observed from collected location fixes with the generated steps created by the step-length and turn-angle distribution generated from the observed steps. The regression model fits a conditional logistic regression with the observed steps and respective randomly generated steps. The generated artefact shows the regression output (coefficient, exp(coefficient), standard-error, z-statistic and p value) and the coefficient plot.
Movement data in Move2 format
Environmental covariate data in raster file (format .tif, .img etc.)
Movement data in Move2 format
SSF_Coef_table.csv with the regression coefficients
SSF_coef_plot.jpeg with the coefficient plot
Step_length_distribution.jpeg shows the distribution of the step-length
Turn_angle_distribution.jpeg shows the distribution of the turn-angle
Raster files: You can upload your own raster file for running the resource selection function analysis. Make sure that they overlap with the tracking data set and are in WGS 84 projection! You can only upload one file, but this one .tif file can contain multiple layers. Note that if you do not upload a file, a fallback raster file of distance to roads in Y2Y region is provided.
Type: Specify how you want to perform the step selection analysis. Type population will generate the random steps based on step-length and turn-angle distribution of all the individuals and type individual will generate random steps based on step-length and turn-angle distribution separately for all the individuals, default is population.