View on GitHub

World-Food-and-Ecosystems

Exercise 2: opening and analyzing the data in QGIS

Now that we have all the data in our folder, we can open QGIS and load the vector (shapefile) of the watersheds, and the two rasterfiles (the biodiversity tif and the NDVI tif files).

Now we get to the core objective of the exercise: what is the link between biodiversity and available vegetation?

To answer this question, we’ll need to decide and simplify (see course 1)

Of course, for the purpose of this exercise, these decisions have already been taken, and are summarized here:

building block decision
Geographic scale Watersheds
temporal scale similar timespans need to be covered and aggregated over sufficiently large timespan
Assumption more primary producers = more available energy, resulting into a higher biodiversity
Dimensions We’ll consider vegetation cover and mammal biodiversity
Dimension description MODIS NDVI (as proxy for vegetation) and mammal species richness by biodiversity.org

Now that we have simplified we can take the average species richness and average NDVI per watershed

Our vector file now has two extra attribute columns: mean NDVI and mean mammal richness.

Can we now visualize this relationship?

** We now have a scatterplot, but the statistical tools to analyse these data in QGIS are limited. So, let’s try to import this data into Rstudio and build a simple statistical regression