Wintertime Infiltration and the Thermal Dynamics of Black Spruce Peatlands within the Boreal Plains
Date
2022-04-08
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
ORCID
0000-0002-1760-798X
Type
Thesis
Degree Level
Doctoral
Abstract
Preliminary investigations (2005-2008) of subsurface temperatures at various ecosites at a Central Alberta research station within the Boreal Plains revealed a tendency for numerical models to predict lower than observed soil temperatures during the wintertime. In addition, black spruce peatlands were observed to have above freezing temperatures throughout most of the winter in the shallow subsurface, while other ecosites would be frozen at similar depths. The focus of this thesis was to determine why black spruce peatlands were warmer during the winter through field observations and numerical simulations, and any impact that may have on the hydrological cycle.
In subsequent investigations the depth of monitoring was increased from 50 cm to 300 cm allowing for a comprehensive thermal profile over a three-year period (2009 to 2011) to be observed at several ecosites. There was a total of eighteen ecological sites, three of each of six types, which included coniferous, deciduous, harvested, burnt, shallow peatlands, and deep peatlands. Observed wintertime temperatures during the three-year period confirmed the preliminary results that the subsurface of deep peatlands would be frozen for significantly less time than both shallow peatlands and upland sites. It was unclear why the shallow subsurface of the deep peatlands barely ever experienced sub-zero temperatures, especially considering that the thermal conductivities of the wet peat should have resulted in a frost depth deeper than observed, even when considering latent heat losses.
From December 2014 to May 2016 an additional field study was conducted at the deep peatlands, with a focus on the winter months and the possible physical and biological effects of black spruce trees and sphagnum moss on winter infiltration and subsurface temperatures. The objective of the additional study was to develop a conceptual model for the thermal dynamics of a black spruce peatland which could explain the warmer subsurface temperatures and the shallower than expected depth of frost. In addition, an algorithm was developed to allow for more accurate simulation of observed subsurface temperatures. The biological effects were quantified by monitoring adjacent living and dead black spruce trunk temperatures, along with the sphagnum moss layer temperatures overlying the peat.
During the 2015 deployment of additional temperature probes at the black spruce peatland sites large amounts of snowmelt from the tree crown were observed to pool around the base of the spruce trees. Snow captured by the black spruce crown completely melted away after a few sunny days. The snowmelt from the crown then dripped and drained towards the base of the black spruce trees, where often a deep “cavity” would exist to the depth of the water table. The installation of moisture content probes within the cavity, loosely filled with peat, revealed increased moisture content following sunny days. Throughout the winter there were also minor increases in observed groundwater elevation following the sunny days as a result of the infiltration of crown drip through the cavity. There was also a very large increase in observed groundwater elevation at the end of March, likely due to the infiltration of upland snowmelt through the unfrozen black spruce cavity. The existence of wintertime infiltration within black spruce peatlands was confirmed through field observations and groundwater measurements.
Data from the additional temperature probes facilitated the development of a conceptual model that could explain why the subsurface did not freeze, and accounting for wintertime infiltration through the tree cavity. Sphagnum moss was observed to be a strong insulator, with temperatures never decreasing below freezing in the underlying peat, even under the tree crown where there was a minimal snowpack. Black spruce trees were also observed to warm the subsurface beginning in April resulting in colder than expected tree trunk temperatures. The conduction of longwave radiation to the root zone from the crown through sap circulation is also a plausible explanation for the circular patterns of snowmelt around the black spruce tree root mats amid a prolonged snowpack observed until at least early May beyond the extent of the black spruce tree’s roots.
Physically based numerical models utilizing the Simultaneous Heat and Water (SHAW) model were then constructed for six of the eighteen ecological sites from the 2009 to 2011 field study. Models were constructed for: two of the deep peatland sites, with black spruce; two of the shallow peatland sites, with lodgepole pine and black spruce; and two of the burnt sites, with lodgepole pine. The burnt sites were subsequently referred to as upland. The numerical models were constructed and calibrated in increasing complexity, beginning with the upland sites, followed by the shallow peatland sites and then the deep peatland sites. The numerical model was able to accurately simulate subsurface temperatures and the observed depth of frost at the pine upland sites without any alteration to the code, and to a lesser degree the shallow, drier, peatland sites. However, the numerical model, as with the previous attempts from the initial investigation, failed to identify a set of parameters that resulted in the successful simulation of the observed subsurface thermal profile at the deep peatland sites. The numerical model also erroneously predicted frost to the groundwater table.
The SHAW numerical model was then updated to calculate the thermal conductivity of the sphagnum moss layer in series instead of parallel. By representing sphagnum moss in series, along with estimating new snowpack thermal conductivity coefficients, the Root Mean Square Error (RMSE) of the wintertime simulated temperatures were substantially reduced for deep peatland vadose zones and were better than those achieved for the upland sites. Using this approach, the wintertime temperatures of the deep peatlands were accurately simulated for the first time at the site, supporting the conceptual model that the insulation provided by sphagnum moss prevents the vadose zone from freezing. The unfrozen vadose zone, along with draining of the crown snowmelt to black spruce tree cavities allowed for wintertime infiltration to occur. The model simulations demonstrate the importance of having a healthy sphagnum moss layer overlying the peat to prevent the subsurface from freezing and inhibiting the infiltration of snowmelt from the black spruce tree cavity. It further highlights the need to better understand and incorporate biologically mediated effects, such as the insulating capacity of live versus dead moss and peat, into physically based models.
Description
Keywords
Black Spruce Peatlands, temperature profiles, thermal conductivity, depth of frost, SHAW
Citation
Degree
Doctor of Philosophy (Ph.D.)
Department
Civil and Geological Engineering
Program
Civil Engineering