Uncovering the understudied role of microtopography and ground cover on evapotranspiration partitioning in high-elevation wetlands in the Canadian Rocky Mountains

Abstract

A warming climate is projected to alter hydrological regimes in high mountain regions, including the Canadian Rocky Mountains. Rivers originating from the eastern slopes of the Canadian Rocky Mountains provide up to 90% of streamflow to downstream users in the Saskatchewan River Basin and have shown significant declines in summer discharge. In mid to late summer, as streamflow gradually decreases while water demand for agriculture, industry, and domestic use remains high, this reduction in streamflow imposes considerable stress on water use for various needs in downstream areas. Wetlands buffer excess water during floods and alleviate water shortage during droughts, making them crucial for sustainable water resource management. With glacial recession, wetlands may become more widespread; however, their hydrological roles are uncertain. Evapotranspiration (ET) represents the total water loss from the wetland surface to the atmosphere through both evaporation and transpiration, which is typically the largest component of the wetland hydrological cycle. Accurately quantifying ET is essential for understanding ecosystem water use patterns, estimating water budgets, and designing sustainable water resource management strategies. The environmental controls of ET in high-elevation wetlands are currently not well understood. The effects of land surface features such as microtopography, bryophytes, and litter covers, commonly found in these ecosystems, on ground surface evaporation—a component of ET—remain insufficiently explored. This thesis investigated these impacts across three typical high-elevation wetland types on the eastern slope of the Canadian Rocky Mountains: a montane fen (Sibbald), a sub-alpine marsh (Burstall) and a sub-alpine wet meadow (Bonsai). Field measurements of controlled evaporation experiments were conducted over various soil substrates and ground cover types, with and without an overstory willow canopy, during July and August 2021. The results showed that microforms and the overstory willow canopy did not exhibit statistically significant direct impacts on ground surface evaporation. Instead, their influences were indirect, affecting soil temperature profiles and below canopy microclimates, which in turn influenced ground surface evaporation. Conversely, ground covers, including litter and bryophytes, significantly impacted on ground surface evaporation, with effects varying by site and involving complex interactions with evaporative demand and water availability. In high-elevation ecosystems, the lack of adequate measurements of energy and water balance components, primarily due to high instrumentation costs and accessibility challenges, hampers understanding of their hydrological processes. While modelling approaches may be convenient and enable the exploration of temporal variability of energy and water fluxes, they require accurate parameterization, and model validation is often unavailable. It remains uncertain whether and how microtopography and ground covers should be integrated into modelling frameworks to enhance the representation of water and energy fluxes of high-elevation wetland ecosystems. This thesis found that the effects of microtopography on ground surface evaporation were satisfactorily modelled by the Penman–Monteith model and a more complex bryophyte layer model in the Atmosphere-Plant Exchange Simulator (APES), based on soil temperature measured in these microtopographical features. The effects of ground covers and overstory canopy on ground surface evaporation were successfully modelled based on the modelling framework of soil evaporative efficiency (SEE) which is the ratio of actual to potential ground surface evaporation. Given that high-elevation ecosystems generally lack sufficient ground measurements for model calibration and validation, this thesis established a simple approach for modelling the SEE of vegetated surfaces over a range of soil substrates and ground cover types, based on the mass fractions of bryophytes and litter. Additionally, a correction method was developed to account for the effects of an overstory willow canopy on SEE. This novel approach is less parameter-intensive compared to conventional methods and potentially widely applicable beyond the three study sites. Both the field experiments and the proposed SEE model formulation illustrate that ground covers significantly influence ground surface evaporation, a key component of ET. This highlights the importance of further studying the effects of ground covers on both ET and its partitioning into evaporation and transpiration at the ecosystem scale. Since field measurements alone are insufficient to fully capture the daily dynamics of ET partitioning, modeling techniques were applied to achieve this goal. To quantify the relative contribution of ground surface evaporation (E) and vascular plant transpiration (T) to wetland ET at a daily scale, part of the newly developed SEE approach, which estimates the surface resistance of bryophytes and litter covered wetland ground surfaces, was incorporated into the widely used Shuttleworth–Wallace (S-W) model. Based on this model, the relative contributions of E and T fluctuated daily in response to meteorological conditions and soil moisture content during the study period. The average daily T/ET ratios for Sibbald, Burstall, and Bonsai were 0.86, 0.46, and 0.61, respectively. When excluding rainy days, the average T/ET ratios were 0.86 for Sibbald, 0.50 for Burstall, and 0.60 for Bonsai. The modified S-W model performed with satisfactory accuracy when compared to field measurements. Assuming that the modified S-W model accurately captures the fundamental mechanisms driving wetland ET partitioning, this model was employed to simulate daily ET partitioning with varying ground cover fractions to investigate the effects of ground covers on ET and T/ET. The results indicated that bryophytes and litter had a minor impact on the magnitude of T/ET, though their influence may be more pronounced in wetlands with an elevated water table. A literature synthesis was further presented to reveal that wetland ET partitioning is strongly controlled by leaf area index (LAI) and moderated by water table depth. However, due to the absence of surface runoff measurements and continuous groundwater level monitoring, the water balance could not be estimated for each study site. Future research should explore potential interaction effects of water table level with microforms and ground covers as well as possible seasonal variations in their effects on wetland ET. Overall, this research reduced uncertainties associated with land surface heterogeneity in estimating ET and its partitioning in high-elevation wetlands, offered new perspectives for developing ET models, and facilitated the evaluation of ecosystem services and hydrological processes in these ecosystems.