Effects of Geologic Heterogeneity on Permafrost Distribution and Catchment Hydrology in Mountain Environments

Changes in surface runoff from permafrost thaw in mountain catchments can be estimated using numerical cryo-hydrogeology models. However, such models can be complex from a numerical standpoint due to the need to simulate transient thermo-hydrologic feedbacks in highly heterogenous geological settings. Models that also seek to quantify water movement and water-budget contributions from ground-ice thaw must further account for changes in water/ice saturation to continually estimate and update the physical properties that control heat and water transfer in the ground (i.e., thermal and hydraulic conductivity) during the model execution. This has important implications for permafrost hydrology modelling efforts in arid mountain watersheds like the High Andes, where water security is threatened by climate change and the role of permafrost in the hydrologic cycle is unclear.

In this contribution the coupled finite element codes TEMP/W and SEEP/W are used to illustrate ground thermal and hydrologic dynamics for different geological scenarios within a hypothetical mountain slope, characteristic of the High Andes at an altitude of up to 6000 m. The 3 km-long, two dimensional cross-sectional model was developed based on a simplified topography, and ground temperatures and climate data collected within the region. In the first scenario, a uniform hydraulic conductivity is applied to the full model domain. A second scenario simulates a case where the hydraulic conductivity of the ground in the upper 200 m is an order of magnitude higher than for the rest of the model (i.e., as in fractured bedrock or unconsolidated sediment). The scenarios were subjected to a 1,000-yr seasonally cyclic climate forcing, followed by 1,000 years of warming superimposed on inter-annual variability at an average warming rate of 4 deg/100 year.

Model experiments show that the applied variations in hydraulic conductivity support vastly different permafrost and ground ice-content distributions under identical climate forcing. Compared to the uniform hydraulic conductivity case, the scenario with high hydraulic conductivity upper layer produces an increase in the heterogeneity of ice-rich permafrost under the stable climate forcing, and a slightly accelerated rate of permafrost thaw under climate warming. Higher recharge and discharge fluxes across the model surface are also predicted for the high hydraulic conductivity scenario.

The divergence in the results is attributed to preferential flow paths that develop near the model surface in the higher hydraulic conductivity case, which in turn leads to increased spatial complexity in advective heat transfer. This can have profound effects on predictive models aiming to estimate rates of permafrost thaw and discharge behaviour under climate warming, and highlights the need for awareness of uncertainties associated with estimated or assumed thermal and hydrologic properties in modelling large mountain catchments.