Performing Spatial Queries (QGIS3)

Spatial queries are core to many types of GIS analysis. Spatial queries allows you to select features in a layer by their spatial relationships (intersect, contain, touch etc.) with features from another layer. In QGIS, this functionality is available via the Select by Location and Extract by Location Processing tools.

Overview of the task

We will be working with 2 data layers for the city of Melbourne, Australia. Given the data layers for the pubs and bars in the city and locations of all metro stations, we want to find out all bars and pubs within 500 meters of a metro station.

Other skills you will learn

  • Choosing an appropriate projection and re-projecting vector data.

  • Creating buffers.

  • Working with the geopackage (.gpkg) data format.

Get the data

City of Melbourne's Open Data Platform provides many GIS-ready datasets for the city.

Download the Metro Train Stations with Accessibility Information dataset by Metro Trains Melbourne. Export the data in the Original format.

../../_images/data110.png

Download the Bars and pubs, with patron capacity dataset by City of Melbourne's Census of Land Use and Employment (CLUE). Export the data as a CSV.

../../_images/data23.png

For convenience, you may directly download a copy of datasets from the link below:

metro_stations_accessbility.zip

Bars_and_pubs__with_patron_capacity

Data Source: [CITYOFMELBOURNE]

Procedure

  1. Locate the metro_stations_accessbility.zip file in the QGIS Browser and expand it. Select the metro_stations_accessbility.shp file and drag it to the canvas. A new layer metro_stations_accessbility will be loaded in the Layers panel.

../../_images/1105.png

  1. The data layer for bars and pubs is in the CSV format. To load it in QGIS, go to Layer ‣ Add Layer ‣ Add Delimited Text Layer.... ( See Importing Spreadsheets or CSV files (QGIS3) for more details on importing CSV files)

../../_images/2100.png

  1. In the Data Source Manager | Delimited Text dialog, browse and select the downloaded Bars_and_pubs__with_patron_capacity.csv file as File name. The X field and Y field columns should be auto selected to x coordinate and y coordinate respectively. Click Add.

../../_images/340.png

  1. You will see a new Bars_and_pubs__with_patron_capacity layer added to the Layers panel. Both of the input layers are in the Geograhpic Coordinate Reference System (CRS) EPSG:4326 WGS84. For performing spatial analysis, it is recommended to use a Projected Coordinate Reference System (CRS). So we will now re-project both the layers to an appropriate regional CRS that minimizes distortions and allows us to work in units of distance such as meters instead of degrees. Go to Processing ‣ Toolbox.

../../_images/418.png

  1. Search and locate the Vector general ‣ Reproject layer tool. Double-click to launch it.

../../_images/519.png

  1. Select Bars_and_pubs__with_patron_capacity as the Input layer. Click the Select CRS button next to Target CRS.

../../_images/619.png

  1. When selecting a projected coordinate system for your analysis, the first place to look is for a regional CRS for the area of interest. For Australia, the Map Grid of Australia (MGA) 2020 is a UTM-based grid system that is used for local and regional mapping. Melbourne falls in the UTM Zone 55, so we can select the GDA 2020 / MGA zone 55 EPSG:7855` CRS.

../../_images/718.png

Catatan

If you are not sure of the a local CRS for the region that you are working in, selecting a CRS for the UTM zone based on the WGS84 datum is a safe choice. You can find out the UTM zone number of your region using UTM Grid Zones of the World.

  1. Next, click the ... button next to Reprojected and select Save to GeoPackage. Geopackage is the recommended open data format spatial data and is the default data exchange format for QGIS3. A single GeoPackage .gpkg file can contain multiple vector and raster layers.

../../_images/818.png

  1. Name the geopackage as spatialquery and click Save.

../../_images/918.png

  1. When prompted for a layer name, enter bars_and_pubs and click OK. Click Run to reproject the layer.

../../_images/1025.png

  1. The window will switch to the Log tab and you will see the algorithm run and create the new output layer bars_and_pubs.

../../_images/1128.png

  1. Now we will reproject the metro_stations_accessbility layer. Switch back to the Paramters tab in the Reproject layer window. Select metro_stations_accessbility as the Input layer. Keep the same Target CRS. Next, click the ... button next to Reprojected and select Save to GeoPackage. Select the same output file spatialquery (Remember that a single geopackage file can contain multiple layers, so we will save the new layer to the same geopackage file). Enter metro_stations as the Layer name. Click Run.

../../_images/1226.png

  1. Back in the main QGIS window, you will see 2 new layers loaded in the Layers panel: bars_and_pubs and metro_stations. You may turn off the visibility for original layers. Now, we are ready to do the spatial query. As we are interested in selecting bars and pubs within 500m of the metro stations, the first step is to create a buffer around the metro stations that represents our search area. Search and locate the Vector geometry ‣ Buffer tool in the Processing Toolbox and double-click to launch it.

../../_images/1324.png

  1. In the Buffer dialog, select metro_stations as the Input layer. Set 500 meters as the Distance. Save the output to the same spatialquery geopackage and enter metro_stations_buffers as the Layer name. Click Run.

../../_images/1422.png

  1. You will see a new metro_stations_buffers layers loaded in the Layers panel. Now we can find out which points from the bars_and_pubs layer falls within the polygons from the metro_stations_buffers layer. Locate the Vector selection ‣ Extract by Location tool from the Processing Toolbox and double-click to launch it.

../../_images/1519.png

Catatan

Extract by location will create a new layer with the matching features from the spatial query. If you just want to select features, use the Select by location tool.

  1. In the Extract by location dialog, select bars_and_pubs as the Extract features from. Check Intersect as the geometry predicate. Set metro_stations_buffers as By comparing to the features from. Save the output to the spatialquery geopackage as the layer selected. Click Run.

../../_images/1618.png

  1. Once the processing finishes, you will see the selected layers added to the Layers panel. Note that this layer only contains points from the bars_and_pubs that fall within the buffer polygons.

../../_images/1719.png

  1. Our analysis is complete. You may notice that the buffer polygons look oval-shaped. This is because our Project CRS is still set to EPSG:4326 WGS84. To better visualize the results, you can go to Project ‣ Properties ‣ CRS and select GDA 2020 / MGA zone 55 EPSG:7855 which we used for the analysis. Once set to this CRS, the buffer will appear in the correct shape.

../../_images/1817.png

If you want to report any issues with this tutorial, please comment below. (requires GitHub account)