Data cross-compatibility
❤️ Make the data work together
Your team will work to make your data more analysis-compatible with the data from the second team you worked with in the previous exercise. This means:
- Data should be in an open format with correct geospatial metadata.
- Data should be in the same format if possible. For example, if both datasets are raster, make both GeoTiff.
- Data should be in the same projection
- Rasters should be co-registered (in the same projection and resolution and orientation)
Other transformations to consider for each dataset:
- Raster <-> Vector
- Subsetting/clipping (e.g., subset one layer to the other’s extent)
- Buffer vector features
- Polygons -> points (find the centroid)
- Dissolve overlapping polygons
- Many more! Explore the processing toolbox tools!
Build this transformation process out in a Jupyter Notebook that guides the reader through each step.
Remember that the choice of what transformations need to be done to a dataset to make it ‘compatible’ with another is often driven by the planned analysis. If you are having trouble coming up with ideas for how to transform data for this exercise, think about how you might use the two datasets together to answer a question.
For example, if you are trying to make a raster containing ice sheet velocity data work together with a vector dataset containing points that represent the location of study sites, you may want to ask what the average ice sheet velocity is within 10km of each point. In order to do this, you would first buffer each point by 10km, and then use the zonal statistics tool to get the average velocity within each resulting polygon.
🗃️ Add the notebook to your team’s GitHub repository
Using the GitHub interface, add this notebook to your team’s repository. Give it a useful name!
⏱️ Finished early?
Please select one of the following data scenarios and develop a Jupyter Notebook to resolve the scenario.