The scale of the hidden reservoir is staggering. Ground ice stored in permafrost worldwide is estimated at approximately 36,600 cubic kilometers, equivalent to five times the annual discharge of the Amazon River. On the Tibetan Plateau alone, permafrost ice content is estimated at 12,700 cubic kilometers, a water equivalent of twelve times the annual discharge of the Yangtze River, of which roughly 3,330 cubic kilometers sits within the shallow permafrost layer between 2 and 10 meters depth. Unlike glacial melt or snowmelt, however, the volume of water released from thawing permafrost is extremely difficult to measure directly, and its contribution to river discharge and groundwater recharge within watersheds remains poorly constrained. This is precisely the gap the review set out to close, arguing that permafrost degradation can alter water circulation through both direct pathways, such as ground ice meltwater release, and indirect ones, such as changed flow paths and connectivity.

To build the synthesis, the researchers followed a transparent systematic protocol. They searched four databases, Web of Science, Scopus, Google Scholar, and the Chinese National Knowledge Infrastructure, covering the period from 1986 to 2025, with particular focus on studies published after 2015. Search terms combined permafrost, active layer, hydrology, runoff, baseflow, thermokarst, surface water-groundwater interaction, and climate change, with no restrictions placed on study outcomes to minimize publication bias. After two-stage screening of titles, abstracts, and full texts, 215 peer-reviewed studies were retained for formal evidence synthesis. Each study was coded for process type, controlling factors, mechanisms, and uncertainties, with a second researcher independently coding a random sample of 20 studies and achieving inter-coder agreement above 85 percent. The team deliberately avoided a formal meta-analysis, citing heterogeneity in study design, metrics, and spatial scale, and instead compiled quantitative ranges where comparable metrics existed.

The review’s first domain concerns surface hydrological processes, where freeze-thaw cycles exert dramatic seasonal control. During the early spring thaw, infiltrating snowmelt refreezes at the shallow thaw front, sharply reducing infiltration capacity and generating saturation-excess overland flow that dominates the early spring hydrograph. Quantitative synthesis shows that saturation-excess overland flow accounts for 66 to 85 percent of total spring runoff, while interflow within the active layer contributes 14 to 34 percent. In summer, runoff becomes intermittent: observations on the Tibetan Plateau show that events typically last less than 24 hours and are often absent when antecedent soil moisture is low. When rainfall intensity exceeds the soil infiltration rate, infiltration-excess, or Hortonian, overland flow can occur locally, especially on sparsely vegetated surfaces. The review also highlights water tracks, unchannelized depressions that concentrate flow down hillslopes, as critical but understudied conduits modulating water and sediment flux, whose development may accelerate under warming.

The second domain treats the active layer as a dynamic hydrological regulator, the critical interface mediating water exchange between the surface and the permafrost. As the thaw front deepens through the season, infiltration capacity rises, shallow groundwater accumulates above the thaw front, and surface runoff coefficients fall. The regulatory strength depends on permafrost type, soil composition, vegetation cover, and topography. Laboratory gas-permeability measurements on sandy soils reveal a striking nonlinear threshold: when volumetric ice content exceeds approximately 40 percent, hydraulic conductivity drops sharply, forcing meltwater toward surface runoff, while below this threshold subsurface flow dominates. The authors caution that such thresholds shift with soil texture and ice distribution and should not be treated as universal constants. Suprapermafrost water, perched immediately above the permafrost table, is central to streamflow generation: observations from permafrost catchments on the Qinghai-Xizang Plateau indicate it can contribute 57 to 66 percent of channel runoff variations, and active-layer water accounts for 59 to 87 percent of ground ice near the permafrost table.

The third domain addresses the hydrological consequences of permafrost thaw, and here the evidence points to a widespread intensification of winter runoff and baseflow. Analyses of runoff changes in Russia found that reductions in freeze depth led to winter runoff increases ranging from 50 to 120 percent, with 56 percent attributed to changes in the freezing front, 38 percent to thawing processes, and 6 percent to increased autumn soil moisture. In the source region of the Yellow River, each 1 meter increase in active layer thickness and thaw depth was associated with winter runoff increases of approximately 150 and 400 cubic meters per second, respectively. In the upper reaches of the Heihe River in the Qilian Mountains, thaw-induced runoff was estimated to contribute roughly 20 percent of mountain outflow, exceeding the proportion from glacial meltwater in that basin. The direct annual contribution of ground ice melt is typically less than 10 percent of total discharge in most basins, but long-term records in the Yukon River Basin show a steady rise in the groundwater fraction of streamflow, consistent with growing connectivity to deeper aquifers.

Thermokarst, the ground collapse that follows ice-rich permafrost thaw, emerges as one of the most visually dramatic manifestations of degradation. In the Tanana Flats of central Alaska, thermokarst activity increased the area of large fens by 26 percent between 1949 and 2018. Conversely, in the headwater area of the Yellow River, the number of water bodies larger than 0.36 hectares decreased by 40 percent, with a 25 percent reduction in total surface area from 1986 to 2015, linked to increased subsurface drainage. Stable isotope analyses and modelling studies have documented significant hydrological connections between precipitation, river water, thermokarst lakes, and permafrost ice melt, with data from 2011 to 2012 showing that ground-ice melt contributed approximately 61 percent of the water in thermokarst lakes on the Tibetan Plateau. Taliks, unfrozen zones that develop beneath lakes and river channels, breach the otherwise impermeable permafrost layer and establish vertical pathways connecting suprapermafrost and subpermafrost aquifers, with subpermafrost flow systems operating on timescales of thousands to tens of thousands of years.

The SAP framework itself organizes these findings around four key coupling points: the seasonal freeze-thaw cycle controlling infiltration versus runoff, the nonlinear threshold behavior of hydraulic conductivity relative to ice content, talik development establishing vertical connectivity, and ground ice serving as a long-term reservoir providing sustained multi-year baseflow support. The framework also resolves an apparent contradiction. Increased storage capacity from active layer thickening might seem to imply reduced discharge, yet observations often show increased baseflow. The explanation is that enhanced connectivity through talik formation transforms previously isolated storage volumes into active flow paths, shifting many systems from a storage-dominated to a flow-dominated regime. Runoff coefficients behave bi-directionally, decreasing during spring thaw as infiltration capacity rises and increasing during autumn freeze as surface freezing restricts infiltration. The authors emphasize that the direction and magnitude of hydrological responses depend on permafrost type, talik continuity, ground-ice content, soil texture, topography, geology, precipitation, evapotranspiration, and vegetation, which is why Arctic lowlands show consistent baseflow increases while mountainous regions on the Tibetan Plateau respond more variably.

Looking forward, the review identifies formidable observational and modelling challenges. Current models are highly sensitive to the parametrization of unfrozen water content, which remains scarce and difficult to upscale; in the Mackenzie Basin model, discrepancies in unfrozen water schemes induce deviations of plus or minus 25 percent in simulated summer peak flows. Abrupt processes such as thermokarst are largely absent from existing models, yet satellite observations indicate they contribute to over 40 percent of land subsidence in the Yangtze River headwaters. Emerging techniques, including distributed temperature sensing, advanced geophysics for three-dimensional mapping of ice content, isotope-enabled models, and InSAR and UAV remote sensing, hold promise for closing critical gaps in subsurface flow, ground ice melt rates, and winter groundwater discharge. Biogeochemical indicators offer a novel pathway: dissolved organic matter concentrations are highest in watersheds with extensive underlying permafrost, while nitrate concentrations peak in catchments with less permafrost. Perhaps the most consequential projection is the transition of the Arctic terrestrial freshwater system from surface-water-dominated to groundwater-dominated, bringing mineral-rich groundwater into rivers and disrupting the hydrological-ecological balance between suprapermafrost groundwater and alpine vegetation, a shift the authors argue demands urgent, integrated investigation.

Subject of Research: Integrating surface-active layer-permafrost hydrological processes: A systematic review and research framework

Article Title: Integrating surface-active layer-permafrost hydrological processes: A systematic review and research framework

Article References: Ding, G., Qin, J., Zhang, S., Yang, B., Cui, J., Wang, F., & Yang, J. (2026). Integrating surface-active layer-permafrost hydrological processes: A systematic review and research framework. Hydrology and Earth System Sciences, 30(20), 6549-6578. https://doi.org/10.5194/hess-30-6549-2026

Image Credits: AI Generated

DOI: 10.5194/hess-30-6549-2026

Keywords: Integrating, surface-active, layer-permafrost, hydrological, processes, systematic, review, research, framework, scientific research