Tutorials wish list

[felixcremer]

What are tutorials that we should include here?
Are there already any geo data tutorials available that could be moved here?

What topics do you miss that touch multiple packages that we could include here?

[evetion]

A bit big, but we should work on geocompj book, and make issues for things that don't work yet. See https://geomcompx.org and the language versions at https://r.geocompx.org/ and https://py.geocompx.org/ (the latter is shorter).

[lupemba]

Idea 1: Interactive data visualization.

• Get a country border polygon from some source like Natural Earth.
• Read in a global data set with multiple variable from a netcdf file.
• Extract only the data within country boarders.
• Visualize the subset using Tyler.jl with a drop down menu to select between variables.

Idea 2: Co-location of data sets.

• Load two partially overlapping datasets from different formats and with different projections (Dataset A and Dataset B).
• Create a co-located dataset where dataset A and B are joined in a Dataframe for the closest match if their distance is under x km.
• Check correlations between dataset A and B.

[felixcremer]

@lupemba what kind of formats do you invision for your idea 2? Is it vector data or do you also think about raster data?

[lupemba]

@lupemba what kind of formats do you invision for your idea 2? Is it vector data or do you also think about raster data?

EDITED: Idea 2 is motivated by my own work where I sometimes collocate measurement between ASCAT and other satellite instruments (Maybe SAR or altimetry). I normally just treat the data as a collection of point measurement for simplicity.
My normal workflow is just

• Select my observation A.
• Find all observations from B at the same time +- some margin.
• Find the nearest neighbor (in space) by brute force.
• If the distance is small enough I store the observation A and B as a collocated observation.
  This is not optimal but work fine for smaller datasets.