Commanding Cartography: Take Control of Faster, More Elegant Workflows from the Command Line

Compelling cartography has never been easier or more abundant. We are inundated with new tools and technologies. All the while, one of the most powerful assets in the cartographer's arsenal is being overlooked: the command line. Using keystrokes to create maps might sound like a task of yesteryear, but I am here to tell you it is a wormhole to the future. Whether you design maps for national parks or newsrooms, the terminal will enable you to supercharge your workflows with speed and elegance. This talk will introduce some old tricks and new tools for designers on any deadline.

Transcript

1. C O M M A N D I N G
   Joshua Stevens
   C A R T O G R A P H Y
   

   View Slide

2. A BLAST FROM THE PAST
   

   View Slide

3. WHERE WE’RE GOING
   The case for the command line
   Some examples from the Earth Observatory
   3 must-know tools + how to apply them
   (and we’ll only scratch the surface!)
   

   View Slide

4. COMMAND LINE?
   MAPS FROM TEXT?
   NO GUI?
   

   View Slide

5. *
   

   View Slide

6. EARTH OBSERVATORY
   Daily, data-driven visuals (maps + satellite imagery)
   earthobservatory.nasa.gov
   

   View Slide

7. Daily, data-driven visuals (maps + satellite imagery)
   14,600 images since 1999
   EARTH OBSERVATORY
   earthobservatory.nasa.gov
   

   View Slide

8. EARTH OBSERVATORY
   Daily, data-driven visuals (maps + satellite imagery)
   14,600 images since 1999
   457 individual stories in 2017
   earthobservatory.nasa.gov
   

   View Slide

9. EARTH OBSERVATORY
   

   View Slide

10. EARTH OBSERVATORY
    0
    50
    100
    150
    200
    250
    2010 2011 2012 2013 2014 2015 2016 2017
    2 images
    3+ images
    15
    30
    45
    60
    75
    2013 2014 2015 2016 2017
    Reporting is Increasingly Visual Visuals are Increasingly Map-only
    

    View Slide

11. WHY BOTHER WITH CLI?
    Some Examples
    

    View Slide

12. View Slide

13. Global data @ 500 m/px
    Consistent data scaling
    Multiple layers (land, water, night lights, +clouds)
    

    View Slide

14. Global data @ 500 m/px
    Consistent data scaling
    Multiple layers (land, water, night lights, +clouds)
    Processed with GDAL
    

    View Slide

15. View Slide

16. New data every 3 hours
    NetCDF format
    Multiple variables
    

    View Slide

17. New data every 3 hours
    NetCDF format
    Multiple variables
    Processed with GDAL
    

    View Slide

18. View Slide

19. Two weeks of hourly data (336 files)
    NetCDF format
    Processed in GDAL
    

    View Slide

20. View Slide

21. High-dimensional data (lat, lon, temp, depth, time)
    
    Map component + temp at depth
    
    Two years of daily measurements (730 frames)
    
    

    View Slide

22. Made with matplotlib + ImageMagick
    High-dimensional data (lat, lon, temp, depth, time)
    
    Map component + temp at depth
    
    Two years of daily measurements (730 frames)
    
    

    View Slide

23. Made with matplotlib + ImageMagick
    High-dimensional data (lat, lon, temp, depth, time)
    
    Map component + temp at depth
    
    Two years of daily measurements (730 frames)
    
    

    View Slide

24. Made with matplotlib + ImageMagick
    High-dimensional data (lat, lon, temp, depth, time)
    
    Map component + temp at depth
    
    Two years of daily measurements (730 frames)
    
    

    View Slide

25. Made with matplotlib + ImageMagick
    High-dimensional data (lat, lon, temp, depth, time)
    
    Map component + temp at depth
    
    Two years of daily measurements (730 frames)
    
    

    View Slide

26. TOOLS YOU NEED TO KNOW
    gdal_translate
    Convert formats
    Scale data (32, 16, 8 bit)
    Georeference
    Resize
    Resample
    Crop
    gdalwarp
    Reproject
    Resample
    gdaldem
    Hillshade
    Slope
    Aspect
    …
    

    View Slide

27. TOOLS YOU NEED TO KNOW
    gdal_translate
    Convert formats
    Scale data (32, 16, 8 bit)
    Georeference
    Resize
    Resample
    Crop
    gdalwarp
    Reproject
    Resample
    gdaldem
    Hillshade
    Slope
    Aspect
    …
    Apply color palettes**
    

    View Slide

28. EXAMPLE: SEA SURFACE TEMP
    https://coralreefwatch.noaa.gov/satellite/bleaching5km/index.php
    $ gdal_translate 'NETCDF:"ct5km_sst_v3.1_20181015.nc":sea_surface_temperature' sst_20181015.tif
    1: Convert NetCDF to GeoTIFF
    

    View Slide

29. https://coralreefwatch.noaa.gov/satellite/bleaching5km/index.php
    $ gdal_translate -projwin -90 70 20 0 sst_20181015.tif sst_20181015_cropped.tif
    2: Crop to an area of interest
    EXAMPLE: SEA SURFACE TEMP
    

    View Slide

30. EXAMPLE: SEA SURFACE TEMP
    https://coralreefwatch.noaa.gov/satellite/bleaching5km/index.php
    $ gdalwarp -t_srs '+proj=aea +lat_1=4 +lat_2=55 +lon_0=-35'
    sst_20181015_cropped.tif sst_20181015_projected.tif
    3: Reproject (Albers Equal Area)
    

    View Slide

31. EXAMPLE: SEA SURFACE TEMP
    https://coralreefwatch.noaa.gov/satellite/bleaching5km/index.php
    $ gdaldem color-relief sst_20181015_projected.tif RdBu.txt sst_20181015_rgb.tif
    4: Apply a color palette
    

    View Slide

32. EXAMPLE: SEA SURFACE TEMP
    https://coralreefwatch.noaa.gov/satellite/bleaching5km/index.php
    4: Apply a color palette
    value R G B
    nv = no data
    $ gdaldem color-relief sst_20181015_projected.tif RdBu.txt sst_20181015_rgb.tif
    

    View Slide

33. ANIMATION IS EASY
    Just do the previous steps a bunch of times
    for f in *.tif; do
    gdaldem color-relief “$f" Colors.txt “${f%.*}_rgb.tif”
    done
    

    View Slide

34. …then make a gif (or video)
    ANIMATION IS EASY
    Just do the previous steps a bunch of times
    for f in *.tif; do
    gdaldem color-relief “$f" Colors.txt “${f%.*}_rgb.tif”
    done
    convert -delay 0 *_rgb.tif animation.gif
    

    View Slide

35. MAKE A MAP SANDWICH
    Base map (GIS, Illustrator, …)
    

    View Slide

36. MAKE A MAP SANDWICH
    Thematic layers (CLI, Python, …)
    Base map (GIS, Illustrator, …)
    

    View Slide

37. MAKE A MAP SANDWICH
    Annotations (Illustrator, PS, …)
    Thematic layers (CLI, Python, …)
    Base map (GIS, Illustrator, …)
    

    View Slide

38. MAKE A MAP SANDWICH
    https://xkcd.com/303/
    (THEN EAT IT… OR BE PRODUCTIVE WITH NEWFOUND TIME)
    

    View Slide

39. Joshua Stevens
    joshua.e.stevens@nasa.gov
    T H A N K Y O U !
    Get in touch:
    @jscarto
    JoshuaStevens.net/contact
    

    View Slide