Assignment 3: GRASS GIS

Assignment 3: GRASS GIS#

Shadow models can be used in urban planning to get information about heat stress. In this assignment you will use GRASS GIS to perform a shadow model simulation and derive the number of shaded benches at different times of day in the city center of Heidelberg.

This assignment should be done and submitted in pairs of 2 students. Not the same ones as for assignment 1 and 2. You may choose with whom to work with until November 15th. Afterwards groups will be assigned randomly. Post your preferred groups in this issue.

Submission deadline: Friday, Nov 29 2024 on GitLab

Deliverables: The final folder/file structure of your repository should look like this.

|__ fossgis_assignment_03
	|__ data
	|__ documentation
		|__ 01_crs.md
		|__ 02_commands_import.md
		|__ 03_command_shadow_areas.md
		|__ 04_shaded_benches.md
		|__ 05_errors_uncertainties.md
	|__ figures
		|__ mapdisplay.jpg
		|__ sun_mask.png
		|__ animation.gif 
	|__ scripts
		|__ 01_raster_import.bat (windows)
		|__ 01_raster_import.sh (linux/mac)
		|__ 02_shadow.bat (windows)
		|__ 02_shadow.sh (linux/mac)
		|__ 03_ohsome_query.txt
		|__ 04_shaded_benches.bat (windows)
		|__ 04_shaded_benches.sh (linux/mac)
		|__ 05_sunmask.py

If you have questions or problems create a new issue on GitLab: https://courses.gistools.geog.uni-heidelberg.de/fossgis/fossgis_assignment_03/-/issues

0. Preparation#

  1. Fork the repository for this assignment and clone your forked repository to your computer.

    The folder ./data contains the following data set:

    • 3D_model_HD.tif: 3D model of buildings in the Heidelberg city center

1. Get started with GRASS GIS#

To get familiar with GRASS GIS you can watch this brief Introduction to GRASS GIS video. If you need further instructions on how to do certain things in GRASS GIS (e.g. create a location), search online for videos or tutorials. Add the useful ones to this issue to share them with other students.

When you create a new location (i.e. a project) in GRASS GIS you need to define a coordinate reference system. You cannot switch as easily as in QGIS.

  1. Find a suitable UTM coordinate references system (CRS) for Heidelberg. Provide the EPSG code of your selected CRS. Add your answer in file ./documentation/01_crs.md.

  2. Start GRASS GIS and set up a new project.

    • Create a new location called “heidelberg” with the CRS you selected before and

    • Create a new mapset called assignment3 and open it.

If you don’t know how to do this, Search for a video online (e.g. youtube) which explains how to do this. Provide a link to the video in this issue and indicate at which minute the relevant part starts if necessary.

2. Import raster data#

In this section you will write a script called 01_raster_import.bat which imports the raster data required for the analysis into the mapset assignment3 in GRASS GIS. Put all commands in this script along with comments which are required to solve the exercises below.

  1. Switch into the mapset assignment3 using the command g.mapset.

  2. Using the command g.region, print the information about the bounds of the current region and the projection of the mapset.

There are two commands you can use to import raster data in GRASS GIS:

  1. Read the GRASS GIS documentation and explain what the difference is between the two commands. Add your answer in file ./documentation/02_commands_import.md.

  2. Import the 3D model 3D_model_HD.tif into the mapset assignment3 as a new layer called height_model using one of these commands. Explain your choice in a comment in the script and paste the command into the script.

  3. Set the region of the mapset to the layer height_model. Paste the command into the script.

  4. Print the current region. Paste the command into the script. Your output should look something like this:

g.region -p
                                                                   
projection: 1 (UTM)
zone:       32
datum:      etrs89
ellipsoid:  grs80
north:      5473211.5024
south:      5472741.5024
west:       477479.9612
east:       478133.9612
nsres:      2
ewres:      2
rows:       235
cols:       327
cells:      76845

Track your changes: Create a commit containing the changes to the script and the documentation files and push it to GitLab.

3. Create shadow animation#

In this section you will write a script called 02_shadow.bat which calculates shaded areas (Figure 1) at different times of day.


Figure 1: Location of benches and shaded areas in Heidelberg.

  1. Find a GRASS GIS command from the raster group (r.) to calculate shadow areas. Name it along with a link to the documentation page. Add your answer in file ./documentation/03_command_shadow_areas.md.

  2. Use the tool you found above to derive the shaded areas on June 21st 2024 every two hours over the course of the day (e.g. 8:00, 10:00, 12:00, 14:00, 16:00, 18:00, 20:00). Paste the commands into the script.

  3. Use the g.gui.animation tool to create an animation of all shadow masks. Save your shadow animation as an animated GIF as ./fossgis_assignment_03/figures/animation.gif.

How to create an animation with the animation tool:

  • Open the animation tool (File –> Animation Tool)

  • Click on Add new animation (top left button)

  • Add the sun masks as a space-time-dataset and the building height raster as a raster dataset

  • Make sure the sun masks are above the building height raster

  • Select a suitable animation speed

Track your changes: Create a commit containing the changes to the script and the documentation files and push it to GitLab.

Note: The methods for the following part of the assignment (Ohsome API queries) will be introduced in the FOSSGIS session on Nov. 20th.#

4. Download benches from OSM#

  1. Use the elements/centroid endpoint of the Ohsome API to download benches in Heidelberg as they were mapped on June 21st 2024. Use a bounding box with the following coordinates: 8.68954,49.40693,8.69858,49.41119. Use the OSM wiki to find a suitable filter for your query. Save the output to ./data/benches.geojson. Save your ohsome query in a file called ./scripts/03_ohsome_query.txt.

Track your changes: Create a commit containing the downloaded data and the Ohsome query and push it to GitLab.

5. Derive number of shaded benches#

In this section you will write a script called 04_shaded__benches.bat which imports the vector data required for the analysis into the mapset assignment3 in GRASS GIS. Put all commands in this script along with comments which are required to solve the exercises below.

There are two commands you can use to import vector data in GRASS GIS:

  1. Import the file benches.geojson into the mapset assignment3 as a new layer called benches. Explain your choice in a comment in the script and paste the command into the script.

  2. Load the layers benches and height_model into the GRASS GIS map display. You don’t need to provide the commands for this. Make a screenshot and save it as ./fossgis_assignment_03/figures/mapdisplay.jpg.

  3. Convert the shaded areas to vector and clip the layer benches to the two shadow layers you created above. Paste the commands into the script.

  4. Use a command from the vector group (v.) to find out how many benches there are in total and how many of the benches were in the shade at 16:00 and at 20:00. Paste the command into the script and add your answer in the file ./documentation/04_shaded_benches.md.

  5. Can you identify any errors or uncertainties in our shadow model? Please explain your thoughts in the file ./documentation/05_errors_uncertainties.md.

Track your changes: Create a commit containing the changes to the script and the documentation files and push it to GitLab.

Bonus: Use Python for the shadow mask calculation#

To create the shadow animation, you created multiple shadow masks. Instead of copying the same command multiple times, it is better to use a for loop. We can achieve this by writing a small Python script. Complete the following Python code snippet and run the commands one by one in the GRASS GIS Python console.

import grass.script as gscript


hours_of_day = [8, 10, 12, 14, 16, 18, 20]
# hours_of_day = list(range(8, 22, 2))

for h in hours_of_day:
    print(h)
    gscript.run_command('r.sunmask', hour=h, ...)

Note: The GRASS GIS Python console will be introduced in the FOSSGIS session on the 27th of November.

Save the script under ./scripts/05_sunmask.py. Now, try loading the script into the GRASS GIS Python editor and running it from there.

Track your changes: Create a commit containing the changes to the script and push it to GitLab.

Resources#

Contents