Permafrost in marine sediments exhibits a lower freezing point and significant unfrozen water content. This paper investigates the role of the soil freezing characteristic curve (SFCC) in permafrost degradation. Three SFCCs, representing thawing-freezing characteristics of soils with varying clay content and salinity, were established based on experiments and existing data. These SFCCs were then applied in numerical analyses to simulate permafrost thawing under various warming scenarios, using measured ground temperatures and permafrost profiles for a site at Longyearbyen in Svalbard (Norway). It is shown that the ground temperature in non-saline permafrost soil increases more rapidly than saline permafrost, due to a greater downward net heat flux to the permafrost in the former case. Conversely, the thawing rate is more pronounced for saline permafrost soil, attributed to its lower freezing point and latent heat consumption. A more nonlinear ice-melting process is observed for permafrost soil with a lower salinity. The temperature rise follows three stages: a constant-rising, a damp-rising, and an accelerated-rising rates. The duration of the damp-rising rate becomes shorter for saline permafrost under a great warming condition. The study underscores the high significance of the soil-freezing characteristic curve for accurate estimations of permafrost degradation.

Frozen soil is a complex four-phase porous medium consisting of soil solid/rock, air, unfrozen/liquid water and ice at the subzero temperatures. Freeze-thaw cycles change the magnitude of total soil water content as well as the unfrozen water/ice ratio in frozen soil that affects soil structure and strength, infiltrability/permeability, water availability for microbial activity and chemical reactions, solute concentration and distribution, and thermodynamics. Accurate quantification of unfrozen water content is therefore critical to understand frozen soil hydrological, biogeochemical, thermal and mechanical properties and processes under climate change. Currently a variety of techniques and methods have been applied to obtain unfrozen water content in frozen soils. However, only few studies have attempted to review and synthesize these works. The objective of this study was therefore to review and collate currently available methods determining unfrozen water content in frozen soils. The principles, applications, advantages and limitations of these methods were reviewed and categorized into five categories: a pressure-based method, radioactive-methods, electromagnetic-methods, thermal-methods, and a sound-based method. Models for indirectly estimating unfrozen water content based on empirical temperature relationships, the soil water/moisture retention characteristic, and the vG-Clapeyron model, were also summarized. There is no direct method to estimate ice content but it can be indirectly calculated based on water balance (i.e., difference between total and unfrozen soil water content). The review is closed with a brief review of future needs and perspectives for simultaneous measurement of unfrozen water and ice contents in the laboratory and in the field.