Terrain (DEM)

TPI: Is This Pixel on a Ridge, a Slope, or in a Valley?

Concept article · Updated · by Dr. Anant Kumar Pathak

The Topographic Position Index answers a deceptively useful question at every point on the map: are you higher or lower than your surroundings? That single comparison is enough to classify a whole landscape into landforms.

The physical basis

TPI subtracts a cell's elevation from the mean elevation of a surrounding neighbourhood. Positive values sit above their surroundings (ridges, peaks); negative values sit below (valleys, pits); near-zero means flat or mid-slope. Changing the neighbourhood radius changes which scale of landform you see.

How it is calculated

TPI = Zcell − mean(Z of the surrounding neighbourhood)

What the numbers mean

Typical range: Centred on 0 (negative = valleys, positive = ridges)

≈ 0 flat or mid-slope · strongly positive = ridge / peak · strongly negative = valley / pit.

Advertisement

The data it needs

Derived fromDigital Elevation Model (SRTM / Copernicus, 30 m) — not an optical band index.

Where it is used

It is used for automated landform classification, habitat and soil mapping, and stratifying terrain for ecological or archaeological survey.

Limitations to know

Entirely scale-dependent — the classification changes with the neighbourhood radius you choose.

TPI is entirely scale-dependent — a small window finds gullies, a large one finds mountain ranges. Report the radius you used, and consider combining two scales to capture nested landforms.

Compute TPI on your own study area

Skip the code. Draw or upload a boundary and Spatial Research Suite runs TPI on live data — with cloud masking, exports and citations built in.

Run this analysis in GISforus →

Frequently asked

What does a positive TPI mean?

The cell is higher than its surroundings — a ridge, crest or peak; negative TPI marks valleys and depressions, and near-zero flat or mid-slope ground.

Why does the neighbourhood size matter for TPI?

TPI describes position relative to its window: a small radius detects fine features like gullies, a large radius detects broad landforms.

Primary reference: Weiss, A. (2001). Topographic position and landforms analysis. ESRI User Conference, San Diego, CA (poster).