Animal Insights Challenge

Examining the hidden lives of animals in Zimbabwe through citizen science
The data consists of counts for each species. For each year, there are four counts, corresponding to morning and afternoon transect walks on Saturday and Sunday. In addition to these counts, we have sourced monthly data for both the park as a whole and the central floodplain on which the actual sampling occurs (the Region of Interest, ‘ROI’). You may use this data to investigate how rainfall, vegetation and water distribution affect the movements of the different species.
The additional data includes:
Climate data from the TerraClimate dataset, accessed through Google Earth Engine. Column names correspond to the different bands detailed here: developers.google.com
Vegetative indices (EVI and NDVI) derived from MODIS imagery, starting in the year 2000.
Surface water distribution, based on the work detailed in this article and made available in the JRC GSW dataset in Google Earth Engine
There may be some gaps in both the counts and the additional data. Of note is that the additional data is provided as monthly figures. Since the game count occurs in September/October, you may wish to focus on periods that are relevant to your investigation.
The starter notebook provides an example to show how the extra data can be combined with the game count figures to test a hypothesis.
Files available for download:
game_counts.csv - the counts for different species for each year from 1995 to 2014 (NB: the 2010 count has corrections made for transects that were not walked that year)
PARK_Additional_Data.csv - monthly climate, vegetation and water data aggregated over the park as a whole
ROI_Additional_Data.csv - monthly climate, vegetation and water data for the region of interest
starter_notebook.ipynb - A brief introduction to the data, showing how to include some of the additional data. Link to Google Colab.