Skip to content

Spatial Queries and Neighborhood Analysis

Unlike attribute selections, spatial queries filter, select, and aggregate datasets based on their geometric relationships—such as proximity, containment, intersection, and overlap. This section details spatial predicates, spatial joins, zonal statistics calculations, and proximity analyses.

We will use the localized Natural Earth vector layers and the digital elevation model grid output_hh.tif:

1. Spatial Relationships (Spatial Predicates)

Spatial predicates are boolean queries evaluating the coordinate overlap between two features. QGIS utilizes standard OGC spatial predicates inside the Select by Location tool (Vector > Research Tools > Select by Location):

    SPATIAL GEOMETRIC RELATIONS
    +-----------------+-----------------+-----------------+
    |   INTERSECTS    |    CONTAINS     |     WITHIN      |
    |   (Any touch)   |   (All inside)  |   (Is inside)   |
    |      O---O      |     [  o  ]     |     o [   ]     |
    +-----------------+-----------------+-----------------+
    |    DISJOINT     |     TOUCHES     |    OVERLAPS     |
    |  (No contact)   |  (Shared edge)  | (Partial cross) |
    |   O   O         |     [ ][ ]      |     [ o ]       |
    +-----------------+-----------------+-----------------+
  • Intersects: Evaluates as True if two geometries share any coordinate point in common (including overlapping areas, crossing lines, or shared edge boundaries). This is the default spatial query.
  • Contains: Feature A contains Feature B if Feature B lies entirely within the boundary of Feature A. For example, a country polygon contains city points.
  • Within: The exact inverse of contains. Feature A is within Feature B if Feature A lies entirely inside the boundary of Feature B.
  • Disjoint: Evaluates as True if two features share absolutely no coordinate space. They are completely separated.
  • Touches: Features touch if they share a border or vertex point, but their internal areas do not overlap. For example, two adjacent sub-catchment polygons touch along their shared watershed divide.
  • Overlaps: Features overlap if they share area space but neither contains the other.

Exercise: Select Cities near Major Rivers (Proximity Selection)

Select all major global cities (ne_10m_populated_places) located near major river buffer corridors.

  1. Load ne_10m_populated_places (point layer) and the major_rivers_buffer_50km polygon layer (created in the previous section).
  2. Go to Vector > Research Tools > Select by Location.
  3. Set the parameters:
  4. Select features from: ne_10m_populated_places.
  5. Where the features (geometric predicate): Check intersect.
  6. By comparing to the features of: major_rivers_buffer_50km.
  7. Modify current selection by: Select creating new selection.
  8. Click Run. The cities situated within \(50\text{ km}\) of major river basins will be highlighted on the map canvas.

2. Spatial Attribute Joins

A Spatial Join links the attribute tables of two layers based on their geographic location rather than a matching text key column.

  • Accessing: Go to Vector > Data Management Tools > Join Attributes by Location.

Exercise 1: One-to-One Spatial Join

Assign country name details to point cities based on the country boundaries they fall within.

  1. Load ne_10m_populated_places (point) and ne_10m_admin_0_countries (polygon).
  2. Open Vector > Data Management Tools > Join Attributes by Location.
  3. Set the parameters:
  4. Input Layer (Target): ne_10m_populated_places.
  5. Join Layer (Source): ne_10m_admin_0_countries.
  6. Geometric predicate: Select within.
  7. Fields to add: Select NAME and CONTINENT.
  8. Join type: Select Take attributes of the first matching feature only (one-to-one).
  9. Joined layer: Save as cities_with_country_names.
  10. Click Run. Verification: Open the output's attribute table to confirm each city now maps its country name.

Exercise 2: Summarized Spatial Join (One-to-Many)

Aggregate city points into country polygons to count how many major cities are in each country and find the maximum city population.

  1. Open Join Attributes by Location.
  2. Set the parameters:
  3. Input Layer: ne_10m_admin_0_countries.
  4. Join Layer: ne_10m_populated_places.
  5. Geometric predicate: Select contains.
  6. Fields to add: Select pop_max (the city population column).
  7. Join type: Select Take summary of matching features (one-to-many).
  8. Summaries to calculate: Check count (gives city counts per country) and max (gives largest city population).
  9. Joined layer: Save as countries_city_summaries.
  10. Click Run. Look at the generated columns pop_max_count and pop_max_max in the output layer.

3. Zonal Statistics (Raster-Vector Overlay)

Zonal Statistics calculates statistical summary metrics (mean, maximum, minimum, standard deviation, sum) of a raster grid within the boundaries defined by a vector polygon layer.

    +----------------------------------+
    |   Raster Grid (e.g. DEM)         |
    |   [2500][2600][2700][2800][2900] |
    +----------------------------------+
         |
         |  ZONAL OVERLAY (State / Province Polygon)
         v
    +----------------------------------+
    | Calculated Vector Output         |
    | dem_mean Elevation  = 2700.0 m   |
    | dem_min Elevation   = 2500.0 m   |
    | dem_max Elevation   = 2900.0 m   |
    +----------------------------------+
  • Accessing: Open the Processing Toolbox and search for Zonal Statistics.

Exercise: Calculate State-Level Elevation Stats

Extract mean, minimum, and maximum elevations for the provinces of Nepal using the output_hh.tif DEM grid and the state boundaries vector.

  1. Load ne_10m_admin_1_states_provinces (polygon) and output_hh.tif (continuous DEM).
  2. Filter the state boundaries to Nepal: Open Select by Expression on the states layer and run: "admin" = 'Nepal'.
  3. Search for Zonal Statistics in the Processing Toolbox.
  4. Set the parameters:
  5. Raster Layer: output_hh.tif (the elevation grid).
  6. Vector Layer containing zones: ne_10m_admin_1_states_provinces (check the box Selected features only).
  7. Output Column Prefix: Enter dem_.
  8. Statistics to calculate: Click ... and check Mean, Min, and Max.
  9. Zonal Statistics: Save the output table as nepal_states_elevation_stats.
  10. Click Run.
  11. Open the attribute table of the output layer. Scroll to the far right to see the columns dem_mean, dem_min, and dem_max showing elevation stats for each Nepal province.

Problem Statement: Elevation Stats for a Digitized Custom Study Area (AOI)

Scenario: A conservation group requires elevation summary statistics (mean, minimum, and maximum elevation) within a custom digitized Area of Interest (AOI) representing a proposed wildlife corridor. Follow these steps to digitize a custom boundary and extract its topographic metrics:

  1. Create a Custom Polygon Layer:
  2. Go to Layer > Create Layer > New Temporary Scratch Layer....
  3. Layer name: Enter study_area_aoi.
  4. Geometry type: Select Polygon.

  5. CRS: Match the projected coordinate system of the DEM (output_hh_utm_30m.tif, EPSG:32645). Click OK.

  6. Digitize the Boundary:

  7. Click the study_area_aoi layer in the Layers panel, then click the Toggle Editing (pencil) icon.
  8. Select the Add Polygon Feature (Ctrl+.) tool.
  9. Click on the map canvas to draw a polygon outlining a custom corridor through the valleys and ridges of the DEM. Right-click to close and complete the polygon.

  10. Click the Save Layer Edits icon and toggle editing mode off.

  11. Calculate Zonal Statistics:

  12. Open the Zonal Statistics tool from the Processing Toolbox.
  13. Set the parameters:
    • Raster Layer: output_hh.tif (or output_hh_utm_30m.tif).
    • Vector Layer containing zones: study_area_aoi (your custom digitized polygon).
    • Output Column Prefix: Enter aoi_.
    • Statistics to calculate: Select Mean, Min, and Max.

  14. Click Run.

  15. Review Results:

  16. Open the attribute table of study_area_aoi. The columns aoi_mean, aoi_min, and aoi_max contain the calculated metrics representing the topography of your custom digitized corridor.

4. Proximity Analysis and Distance Queries

Neighborhood analysis evaluates the geographic distance separating features across different layers.

  • Distance Matrix: Calculates the exact metric distance from each feature in a point layer to the nearest features in another layer.
  • Accessing: Go to Vector > Analysis Tools > Distance Matrix.
  • Distance to Nearest Hub (Hub Distance): Calculates the distance between features in an input layer and the closest feature in a "destination" layer, appending the distance metric directly to the input attribute table.

Exercise: Distance to Nearest Major River

Calculate the distance from every global city (ne_10m_populated_places) to the nearest major river centerline.

  1. Load ne_10m_populated_places (point) and ne_10m_rivers_lake_centerlines (line).
  2. In the Processing Toolbox, search for the Distance to nearest hub (points) tool.
  3. Set the parameters:
  4. Source points layer: ne_10m_populated_places.
  5. Destination hubs layer: ne_10m_rivers_lake_centerlines.
  6. Hub layer name attribute: Select name (attaches the name of the nearest river to the city).
  7. Measurement unit: Select Kilometers (or Meters if using a projected system).
  8. Hub distance: Save the output point layer as cities_river_proximity.
  9. Click Run. Open the attribute table of the output layer. You will find the fields HubName (river name) and HubDist (distance to river) appended to every city point.