Jon Warland


DOI bib
Implications of measurement metrics on soil freezing curves: A simulation of freeze–thaw hysteresis
Renato Pardo Lara, Aaron Berg, Jon Warland, Gary Parkin
Hydrological Processes, Volume 35, Issue 7

Soil freeze-thaw events have important implications for water resources, flood risk, land productivity, and climate change. A property of these phenomena is the relationship between unfrozen water content and sub-freezing temperature, known as the soil freezing characteristic curve (SFC). It is documented that this relationship exhibits hysteretic behaviour when frozen soil thaws, leading to the definition of the soil thawing characteristic curve (STC). Although explanations have been given for SFC/STC hysteresis, the effect that “scale”—particularly “measurement scale”—may have on these curves has received little attention. The most commonly used measurement scale metric is the “grain” or “support,” which is the spatial (or temporal) unit within which the measured variable is integrated—in this case, the soil volume sampled. We show (1) measurement support can influence the range and shape of the SFC and (2) hysteresis can be, at least partially, attributed to the support and location of the measurements comprising the SFC/STC. We simulated lab measured temperature, volumetric water content (VWC), and permittivity from soil samples undergoing freeze-thaw transitions using Hydrus-1D and a modified Dobson permittivity model. To assess the effect of measurement support and location on SFC/STC, we masked the simulated temperature and VWC/permittivity extent to match the instrument’s grain and location. By creating a detailed simulation of the intra- and inter-grain variability associated with the penetration of a freezing front, we demonstrate how measurement support and location can influence the temperature range over which water freezing events are captured. We show it is possible to simulate hysteresis in homogenous media with purely geometric considerations, suggesting that SFC/STC hysteresis may be more of an apparent phenomenon than mechanistically real. Lastly, we develop an understanding of how the location and support of soil temperature and VWC/permittivity measurements influence the temperature range over which water freezing events are captured.


DOI bib
In Situ Estimates of Freezing/Melting Point Depression in Agricultural Soils Using Permittivity and Temperature Measurements
Renato Pardo Lara, Aaron Berg, Jon Warland, Erica Tetlock
Water Resources Research, Volume 56, Issue 5

We present a method to characterize soil moisture freeze‐thaw events and freezing/melting point depression using permittivity and temperature measurements, readily available from in situ sources. In cold regions soil freeze‐thaw processes play a critical role in the surface energy and water balance, with implications ranging from agricultural yields to natural disasters. Although monitoring of the soil moisture phase state is of critical importance, there is an inability to interpret soil moisture instrumentation in frozen conditions. To address this gap, we investigated the freeze‐thaw response of a widely used soil moisture probe, the HydraProbe, in the laboratory. Soil freezing curves (SFCs) and soil thawing curves (STCs) were identified using the relationship between soil permittivity and temperature. The permittivity SFC/STC was fit using a logistic growth model to estimate the freezing/melting point depression (Tf/m) and its spread (s). Laboratory results showed that the fitting routine requires permittivity changes greater than 3.8 to provide robust estimates and suggested that a temperature bias is inherent in horizontally placed HydraProbes. We tested the method using field measurements collected over the last 7 years from the Environment and Climate Change Canada and the University of Guelph's Kenaston Soil Moisture Network in Saskatchewan, Canada. By dividing the time series into freeze‐thaw events and then into individual transitions, the permittivity SFC/STC was identified. The freezing and melting point depression for the network was estimated as Tf/m = − 0.35 ± 0.2,with Tf = − 0.41 ± 0.22 °C and Tm = − 0.29 ± 0.16 °C, respectively.