Comparison of algorithms and parameterisations for infiltration into organic-covered permafrost soils

Abstract

Infiltration into frozen and unfrozen soils is critical in hydrology, controlling active layer soil water dynamics and influencing runoff. Few Land Surface Models (LSMs) and Hydrological Models (HMs) have been developed, adapted or tested for frozen conditions and permafrost soils. Considering the vast geographical area influenced by freeze/thaw processes and permafrost, and the rapid environmental change observed worldwide in these regions, a need exists to improve models to better represent their hydrology.

In this study, various infiltration algorithms and parameterisation methods, which are commonly employed in current LSMs and HMs were tested against detailed measurements at three sites in Canada's discontinuous permafrost region with organic soil depths ranging from 0.02 to 3 m. Field data from two consecutive years were used to calibrate and evaluate the infiltration algorithms and parameterisations. Important conclusions include: (1) the single most important factor that controls the infiltration at permafrost sites is ground thaw depth, (2) differences among the simulated infiltration by different algorithms and parameterisations were only found when the ground was frozen or during the initial fast thawing stages, but not after ground thaw reaches a critical depth of 15 to 30 cm, (3) despite similarities in simulated total infiltration after ground thaw reaches the critical depth, the choice of algorithm influenced the distribution of water among the soil layers, and (4) the ice impedance factor for hydraulic conductivity, which is commonly used in LSMs and HMs, may not be necessary once the water potential driven frozen soil parameterisation is employed. Results from this work provide guidelines that can be directly implemented in LSMs and HMs to improve their application in organic covered permafrost soils.

[A & I plus PDF only]

COPYRIGHT: © Author(s) 2010. This work is distributed under the Creative Commons Attribution 3.0 License.