Fresh scarp vs old scarp: what satellite imagery actually shows you
After a storm or a shaking event, half the work of compiling an inventory isn't finding scarps. It's deciding which ones are new. A slope can carry a dozen old failures, a fresh one from last week, and a half-dozen reactivations in between, and from a single image they can look nearly identical at first glance. Here's what separates them.
What changes on a scarp as it weathers
A fresh scarp exposes material that hasn't seen sun, rain, or oxidation yet. In RGB imagery that usually reads as a bright, high-contrast patch against the surrounding slope: pale soil, raw rock, or subsoil color that doesn't match the vegetated ground around it. The crown edge is sharp, almost a clean line where the headscarp cuts the original surface. Talus and debris in the runout path still sit loose and angular, with distinct individual blocks visible if the resolution allows it.
Weathering starts working on that exposed face almost immediately. Oxidation darkens iron-bearing soils within months. Rain washes fines off the surface and starts rounding the crown. Rockfall from the headwall buries the sharpest break in slope. By a year or two out, the color contrast that made the scarp jump out of the image has usually faded into something closer to the background tone, and that's before vegetation gets involved at all.
Vegetation regrowth as a dating clue
Bare ground doesn't stay bare. Pioneer grasses and forbs colonize a fresh scarp face within one to two growing seasons in most temperate settings, faster on moist aspects, slower on steep sun-baked ones. That green-up shows as rising NDVI or just visible patchiness in later imagery, and it's one of the more reliable relative-age markers you have when you're working from imagery alone rather than field notes.
The catch is that regrowth rate isn't a constant you can plug into a formula. A north-facing scarp in a wet catchment can green up twice as fast as a south-facing one in the same drainage. Soil type matters, grazing pressure matters, and in arid terrain a scarp can stay essentially bare for years. Vegetation cover sorts scarps into fresh, recent, and old relative to each other within a region. It's not a stopwatch.
Weathered scarp vs fresh scarp: the full checklist
Color and green-up are the two things that jump out first, but a few other tells help confirm the call:
- Crown edge definition. Sharp and linear means fresh. Rounded, slumped, or softened into the hillslope means it's had time to degrade.
- Debris texture. Angular, loose-looking material with visible individual clasts reads fresh. A smoothed, consolidated apron with drainage channels already cut into it reads old.
- Shadow behavior. In imagery with a decent sun angle, a fresh scarp throws a crisp shadow line at the headwall. An older one has lost enough relief that the shadow softens or disappears.
- Context against known event timing. If you have a before/after pair bracketing a specific storm or earthquake, that's your hardest evidence. Anything that appears in the after-image and wasn't there before is fresh by definition, no interpretation needed.
That last point is the real answer to "how do you date a landslide scarp" from imagery alone: you bracket it rather than measure it. A scarp absent in the before-image and present in the after-image is fresh as of that event, full stop. Cosmogenic exposure dating in the field puts an actual year count on a scarp. Satellite imagery puts a window on it instead: absent, then present, then vegetated by some later date. Stack enough image pairs across a season and that window gets tight, even without ground truth.
The practical problem for anyone compiling a regional inventory is that this comparison work doesn't scale well when it's done tile by tile, valley by valley, across a whole affected region after an event. Landslide Detection builds the region-wide scarp and runout layer from high-res imagery so a survey team gets the full picture of what's fresh across the area at once, instead of reconstructing it drainage by drainage.
If you're staring down a post-event region and need the fresh scarps separated from the old ones before you plan fieldwork, that's the problem this exists to solve.