Lab 10

Overview¶

This lab introduces you to GeoLibre, a free and open-source cloud-native GIS platform that runs in the browser, on the desktop, and inside Jupyter notebooks. You will use the geolibre Python package, which embeds the complete GeoLibre app in a notebook cell and gives you a leafmap-style Python API for driving it.

What makes GeoLibre different from the mapping libraries you have used so far is that state syncs both ways. Data you add from Python appears in the app UI, and edits you make in the UI — panning, zooming, toggling layers, restyling — are readable back from Python. The whole map is a single .geolibre.json project file that is interchangeable between the notebook, the web app, and the desktop app.

Objectives¶

By completing this lab, you will be able to:

1. Create and configure a GeoLibre map in a Jupyter notebook and control its camera from Python.

2. Add vector data from several sources and formats, including GeoJSON, CSV, and GeoParquet.

3. Apply data-driven symbology and annotate a map with legends and colorbars.

4. Add raster layers and compare two of them with a swipe control.

5. Build a 3D visualization using building-footprint extrusions.

6. Read live map state back from the UI and save a project for sharing.

# %pip install geolibre

Exercise 1: Creating Your First GeoLibre Map¶

• Create a map centered on a city of your choice with an appropriate zoom level, using the liberty basemap and a height of 700px. Display the map.

• In a second cell, switch the basemap to dark and move the view to a different city using set_center(). Re-run and confirm the displayed map above updates in place without being recreated.

• Print the list of available basemap names using geolibre.basemap_catalog().

Exercise 2: Adding Vector Data¶

Create a new map and add each of the following layers with an appropriate name and style:

• GeoJSON from a URL: world cities as red circles.

  • https://github.com/opengeos/datasets/releases/download/world/world_cities.geojson

• CSV point data: the same cities from a CSV, using add_csv() with the longitude and latitude columns.

  • https://github.com/opengeos/datasets/releases/download/world/world_cities.csv

• GeoParquet: US states, drawn with a semi-transparent fill so the cities stay visible.

  • https://github.com/opengeos/datasets/releases/download/us/us_states.parquet

Then frame the continental US using set_center() and set_zoom(). (GeoLibre also has zoom_to_layer(), but it asks the live widget for the layer’s extent, so it only works while the map is displayed and connected.)

Exercise 3: Marker Clusters and Heatmaps¶

• Create a map and add the world cities dataset as a marker cluster. Experiment with the cluster_radius and cluster_max_zoom parameters and describe in a Markdown cell what each one changes.

• Create a second map and add the same dataset as a heatmap. Adjust radius and intensity until the population centers read clearly at a continental zoom level.

Exercise 4: Choropleth Symbology and Legends¶

• Create a map and add the US counties dataset as a choropleth using add_choropleth(), classified on the CENSUSAREA column.

  • https://github.com/opengeos/datasets/releases/download/us/us_counties.geojson

• Use a quantile classification with 6 classes and a color ramp of your choice. Then create the same map with an equal-interval classification and, in a Markdown cell, explain why the two maps look so different for this variable.

• Add a colorbar with a meaningful label and units.

Exercise 5: Raster Layers and Swipe Comparison¶

• Thematic raster: Create a map and add the NLCD 2021 land cover raster with the built-in nlcd legend.

  • https://github.com/opengeos/datasets/releases/download/raster/nlcd_2021_land_cover_90m.tif

• DEM: Create a map and add the DEM below with the terrain colormap and an elevation colorbar.

  • https://github.com/opengeos/datasets/releases/download/raster/dem.tif

• Swipe comparison: Create a map centered on Derna, Libya, add the two images below, and connect them with split_map() so the slider compares before and after the September 2023 flood.

  • Before: https://github.com/opengeos/datasets/releases/download/raster/Libya-2023-07-01.tif

  • After: https://github.com/opengeos/datasets/releases/download/raster/Libya-2023-09-13.tif

Exercise 6: 3D Building Extrusions¶

• Create a map centered on lower Manhattan with the dark basemap and add the NYC buildings dataset below.

  • https://github.com/opengeos/datasets/releases/download/places/nyc_buildings.geojson

• Enable extrusion and drive the building heights from the height_avg property. Set the pitch and bearing so the 3D effect is clearly visible.

• Retrieve the layer with find_layer() and use the layer handle to lower its opacity and change the extrusion color.

Exercise 7: Reading State Back from the UI¶

• Using the 3D map from Exercise 6, pan and zoom the map in the UI, toggle a layer, then run a cell that calls to_project() and prints the live center, zoom, and layer names. Run this cell by hand after you have moved the map; a plain Restart and run all executes it before you have touched anything and will just report the values you set in code.

• Call describe() and print the summary. In a Markdown cell, note the feature count reported for the buildings layer.

• Use column() on the buildings layer handle to pull the height_avg values, then report the minimum, maximum, and mean building height. You may use NumPy for this.

Exercise 8: Saving and Sharing a Project¶

• Save your Exercise 6 map to lab10.geolibre.json with save_project().

• Create a new, empty Map and load that file with load_project(). Display it and confirm it matches the map you saved.

• Open https://web.geolibre.app in a browser tab, load the same .geolibre.json file from the app’s File menu, and confirm the project opens there too. Describe what you see in a Markdown cell.

Exercise 9: Geoprocessing in the Browser (Bonus, 10 points)¶

GeoLibre ships more than 1,000 Whitebox geoprocessing tools that run entirely in the browser through WebAssembly. These exercises drive the live widget, so the map must be displayed and finished loading before the tool call runs. Put the Map in one cell, display it, wait for it to render, then run the tool in the next cell.

• Create a map, display it, and print how many tools list_whitebox_tools() reports.

• Add the DEM from Exercise 5 as a raster layer, then run the slope tool against it with units set to degrees.

• Add the resulting slope layer to the map with an appropriate colormap and colorbar.