Distance to habitat — Task

Implementation: Distance to habitat (Recipe - QGIS).

Distance to habitat is the Task of describing, for every place in a study area, how far it is to the nearest habitat patch. It shows where the landscape is open and empty of nature, and it is the basis for questions such as “how much farmland lies more than 200 m from any habitat?”.

Used in examples: Habitat network at Ledreborg (Example).

Task Schema (tool-agnostic)

1. Start from the habitat patches — normally the nature content layer, including its margin.
2. Choose the cell size (decision point) — it sets the precision of the distances.
3. Turn the patches into a raster (Operator: rasterize) — habitat cells 1, all others 0, over the study area plus its margin.
4. Compute the distance surface (Operator: distance surface) — every cell’s distance to the nearest habitat cell.
5. Cut to the study area (Operator: clip to extent) — after the distances are computed, so habitat outside the area still counts.
6. Summarise (Operator: zonal statistics) — e.g. the share of the area more than 100 m from habitat, or the mean distance per landscape.

Output: a distance raster in metres, and summaries per study area or landscape.

Epistemic Choices to record

• Cell size and narrow features: with 10 m cells, a 5 m hedge only becomes a habitat cell if it covers the cell’s centre. Rasterize with all touched (any cell the patch touches) or use smaller cells; at Ledreborg all touched added 21 % more habitat cells and lowered the share of land more than 100 m from habitat from 22 % to 19 %.
• Straight-line distance: distance as the crow flies ignores roads, fields and other barriers. For a species that cannot cross a motorway, a cost-distance would be needed.
• Margin: without habitat outside the study area, places near its edge look far from habitat.
• Classes: distance classes (50, 100, 200 m) should come from the question, e.g. how far a species moves between patches.