What terrain constraints matter most for infrastructure siting?

Steep slope, flood hazard, landslide susceptibility, poor drainage and wet ground, weak or expansive soils, and regulatory exclusions such as protected areas and setbacks. Each is mapped as a layer and combined into a composite constraints surface.

How do you build a constraints map in GIS?

Classify each constraint layer into pass, caution, and exclude, reproject everything to a common metric CRS and grid, then overlay. Hard constraints become no-go masks; soft constraints are weighted and summed into a suitability or constraint score.

What is the difference between a hard and a soft constraint?

A hard constraint is an absolute exclusion — a regulatory no-build zone, a floodway, or an active landslide — that removes the area from consideration. A soft constraint raises cost or risk (moderate slope, wet ground) but can be engineered around.

Which official datasets define flood and landslide constraints?

In the US, FEMA Flood Insurance Rate Maps define regulatory flood zones and USGS provides landslide inventory and susceptibility data. Many countries publish equivalent national flood and geohazard datasets; always prefer the official regulatory source.

Terrain Constraints for Infrastructure Planning

What a constraints analysis answers

Before a site is chosen or a corridor refined, planners need to know where infrastructure cannot go and where it can only go at a cost. A terrain constraints analysis maps those answers: it assembles the physical and regulatory factors that limit development — slope, flood hazard, landslide susceptibility, drainage, sensitive ground, and protected zones — classifies each as a hard exclusion or a soft penalty, and overlays them into a composite picture. The result steers siting away from no-go ground and toward the least-constrained options, while making the basis of every exclusion explicit and auditable. This is screening, not permitting; high-consequence decisions still need specialist assessment.

Hard versus soft constraints

The first discipline is separating the two:

Hard constraints are absolute. A regulatory floodway, a designated conservation area, a statutory setback from a watercourse, or an active landslide body removes the land from consideration regardless of engineering. These become a binary no-build mask.
Soft constraints raise cost or risk but can be mitigated. Moderate slope, seasonally wet ground, moderate landslide susceptibility, or long haul distances are weighted and combined into a graded score.

Conflating the two is the classic error: a soft constraint dressed up as an exclusion sterilises usable land, while a hard constraint treated as merely "high cost" exposes the project to legal or safety failure.

The core constraint layers

Slope

Steep ground raises earthworks, foundation, and stability costs and eventually becomes a hard limit. Derive slope from a DEM reprojected to a metric CRS:

gdalwarp -t_srs EPSG:32633 -tr 30 30 -r bilinear dem.tif dem_utm.tif
gdaldem slope dem_utm.tif slope_deg.tif -alg Horn

Classify into bands appropriate to the asset (for example, for general building: 0–8 degrees favourable, 8–15 caution, >15 generally excluded), but set the breakpoints from the relevant design or planning standard rather than generically.

Flood hazard

Use the official regulatory product. In the US that is the FEMA Flood Insurance Rate Map, which defines Special Flood Hazard Areas (the 1 percent annual chance, "100-year" floodplain) and the floodway. The floodway is typically a hard constraint; the wider Special Flood Hazard Area is a strong soft constraint requiring elevation or flood-proofing. Many other jurisdictions publish equivalent national flood maps; prefer the statutory source over modelled approximations for any regulatory decision.

Landslide susceptibility

Combine observed inventory and modelled susceptibility, and keep them distinct. A mapped active landslide is a hard constraint; a high-susceptibility class from a model is a soft constraint flagging the need for geotechnical investigation. USGS landslide inventory and susceptibility products are a good reference for how agencies present these layers and their uncertainty.

Drainage and wet ground

Poor drainage signals foundation and serviceability problems. Derive flow accumulation and a topographic wetness index from the DEM (GRASS r.watershed, or QGIS Topographic Wetness Index) to flag persistently wet hollows and concentrated flow lines that should be avoided or engineered.

Regulatory and environmental exclusions

Protected areas, heritage zones, buffer setbacks from rivers, wetlands, and habitat designations are usually hard constraints defined by official boundaries. Buffer linear features (ST_Buffer in PostGIS, or QGIS Buffer) to the statutory setback distance before overlaying.

Building the composite overlay

Normalise geometry. Reproject every layer to one metric CRS and resample rasters to a common grid (e.g. 30 m). Mismatched CRS or cell size is the most common source of misaligned overlays.
Classify each layer into pass / caution / exclude with documented breakpoints.
Apply hard constraints as a mask. Union all hard-exclusion polygons/rasters into a single no-go layer.
Weight and sum soft constraints into a constraint score using the QGIS Raster calculator or weighted-overlay, with weights agreed and recorded with the project team.
Mask the score by the hard-constraint layer so excluded ground reads as no-go, not merely "high cost".

A PostGIS pattern for the vector exclusions:

CREATE TABLE no_build AS
SELECT ST_Union(geom) AS geom FROM (
  SELECT geom FROM floodway
  UNION ALL SELECT ST_Buffer(geom, 30) FROM watercourses
  UNION ALL SELECT geom FROM protected_areas
) s;

Worked example: screening a substation site

Reproject the regional DEM and derive slope and topographic wetness index.
Pull FEMA floodway and Special Flood Hazard Area, the protected-area boundaries, and a 30 m watercourse buffer.
Hard mask = floodway ∪ protected areas ∪ watercourse buffer ∪ slope > 15 degrees.
Soft score = weighted sum of (slope band, flood-fringe presence, wetness index, landslide susceptibility class).
Mask the score by the hard layer; rank candidate parcels by mean soft score within the buildable area.
Carry the top candidates to geotechnical and environmental review.

Common pitfalls and why they happen

Mixing CRS and resolution before overlay — produces silent spatial offsets, so an exclusion lands on the wrong parcel.
Treating a model as a regulation — using a modelled flood layer where the statutory FIRM governs invites a permitting failure.
Undocumented weights — stakeholders cannot contest a constraint score whose weights are hidden.
Ignoring data age and scale — a 1:50,000 hazard layer cannot adjudicate a parcel-scale boundary; flag the mismatch.
Collapsing inventory and susceptibility — losing the distinction between an observed landslide and a modelled "could happen here".

QA and validation

Confirm all layers share CRS, extent, and grid before overlay (gdalinfo, layer properties).
Spot-check exclusions against aerial imagery and known site conditions.
Verify the hard mask is genuinely binary and that the soft score is reported on a defined scale with a legend.
Record the source, date, scale, and authority of every constraint layer in the report.

Bathyl perspective

We build constraints maps to make siting decisions defensible: hard exclusions traceable to a named regulation or observed hazard, soft penalties carrying their weights and sources in the open. The map narrows the field and tells you where deeper investigation is non-negotiable, without ever standing in for the specialist who signs it off.