Vegetation Recovery Following Retrogressive Thaw Slumps Across Northern Tundra Regions
- Aug 12
- 3 min read
RTS recovery varies from a decade in low-Arctic tundra to over a century on the Tibetan Plateau — and regional GPP predicts the difference.
Retrogressive thaw slumps (RTS) — mass-wasting collapses triggered by thawing ice-rich permafrost — are accelerating across the Arctic, with activity up sixty-fold on Banks Island since the 1980s and threefold on the Qinghai–Tibet Plateau in just the past several years. Each slump strips vegetation and mobilizes long-buried soil carbon, but whether and how fast that vegetation grows back has remained poorly quantified at scale.
Xia et al. (2026) close that gap, combining nine years of PlanetScope satellite imagery with historical aerial photos and drone surveys across 12 slump-affected regions spanning Alaska, Canada, Siberia, and the Tibetan Plateau to build the first regionally scalable model of thaw-slump recovery.
Key Findings
Recovery timelines split sharply between low-Arctic and high-Arctic/high-elevation sites
In the low-Arctic regions of Toolik Lake, Noatak, the Mackenzie Delta, and Yamal, surface greenness (measured via NDVI) rebounded to pre-disturbance levels in just 5.4 to 9.3 years, after which it stayed stable with only slight further increases. In the high-Arctic (Banks Island) and high-elevation Tibetan Plateau sites, the same process took 34.6 to 105 years, with the two slowest-recovering regions — western Qinghai-Tibet and eastern Banks Island — showing such sparse vegetation and limited observation windows that full recovery may run even longer than estimated.
A single productivity metric predicts recovery time across all regions
Rather than climate variables, the strongest predictor of recovery time turned out to be regional gross primary productivity (GPP), derived from satellite-based solar-induced chlorophyll fluorescence. Recovery time follows a tight power-law relationship with GPP (τ = 1.35 × GPP⁻¹·⁶⁸, R² = 0.97) — ecosystems with higher productivity recover dramatically faster. The model was validated against four independent sites in Siberia and Canada, where predicted recovery times (3.6 to 24.1 years) closely matched recovery times measured directly from historical high-resolution imagery (4 to 25 years). Notably, temperature and precipitation alone showed no strong relationship with recovery time, suggesting GPP captures the combined effect of climate, nutrient availability, and moisture more reliably than any single climate variable.
Spectral recovery masks a deeper story of vegetation succession
A returned NDVI value doesn't mean the same plant community came back. In low-Arctic sites, low-stature vegetation — mosses, graminoids, dwarf shrubs — colonized disturbed ground within 5–10 years, then was progressively replaced over the following decades by taller erect shrubs, producing surfaces that were sometimes greener than the surrounding undisturbed tundra. On the Tibetan Plateau, by contrast, graminoids and forbs continued to dominate recovered scars even after 50 years, likely due to a scarcity of nearby shrub seed sources and harsher growing conditions. The authors frame this as a genuine state change — tussock-shrub tundra transitioning into shrub tundra in some regions while staying locked in a low-productivity state in others.
Recovery in Context
Placed alongside other Arctic disturbances, RTS recovery is notably slow. Drained thermokarst lake basins re-vegetate within about 2 years, and postfire tundra recovers within roughly 10–13 years in both boreal Siberia and Alaska — comparable to or faster than even the quickest RTS recovery observed here. That gap matters for carbon accounting: RTS mobilize previously frozen soil carbon and increase microbial decomposition on release, so the pace at which vegetation re-establishes shapes how much of that carbon loss gets offset by renewed uptake. In productive low-Arctic settings, the eventual shift to shrub cover may even create a modest negative feedback that enhances long-term carbon sequestration — but in the Arctic's least productive, coldest terrain, disturbed ground can stay a net carbon source for a century or more.
A Slow-Motion Reshaping of the Tundra
RTS are geographically constrained — they need ice-rich permafrost and specific slope and soil conditions — but the areas that meet those conditions are exactly the ones warming fastest. As slump activity continues to expand across permafrost landscapes, Xia et al.'s productivity-based framework offers something rare in this space: a way to forecast, region by region, whether a given scar is on track to green within a decade or remain a bare, carbon-losing wound in the landscape for generations.
Sign up for our newsletter or connect with us on social media to stay up-to-date with our latest posts and permaculture inspiration.
Explore our inspiring series and posts:
Love the post? Share it with your circle, inspire your people.