Abstract
The glaciers in the McMurdo Dry Valleys (MDV) melt very little. Glacial melt is the
main source of water for the streams that transport freshwater and nutrients to the
terminal lakes. In this harsh environment, a microbial ecosystem relies on the freshwater
for survival. Currently, MDV glaciers are in stable equilibrium and melt very
little compared to other alpine glaciers. However, the melt regime currently rests
on the threshold of change in this warming world. It is anticipated that changes to
glacial melt will affect the survival and biogeography of the microbial communities.
Thus, a key challenge is to resolve the atmospheric controls on glacial melt and
streamflow to understand how future climate may impact hydrological connectivity.
To do so, a novel hydrometeorological modelling system was adapted to simulate
cold-based glacial melt on a MDV glacier. The modification of the meltwater percolation
scheme to allow near-surface runoff enabled the non-negligible contribution of
small-scale melt processes to seasonal runoff to be resolved. In addition, the spectral
albedo scheme parameters were optimized to simulate snow-albedo-melt feedbacks
for the first time on a MDV glacier. Accurately simulating albedo is vital, as net
shortwave radiation is the primary driver of melt in this energy-limited environment.
Adapting and testing the hydrometeorological modelling system at a point-scale laid
the foundation to model spatially distributed mass balance and streamflow in Taylor
Valley. This is the first modelling system capable of simulating sub-diurnal scale
streamflow variability at ecologically relevant timescales. By bridging glacial melt
to streamflow, the differences in the meteorological drivers on streamflow over two
seasons are explained. Furthermore, it was found that the second key driver of melt,
the sensible heat flux, can generate a similar amount of total seasonal streamflow
over a single event to a season driven by frequent, smaller, net shortwave radiation
driven melt events. The application of the modelling system was extended to investigate
the sensitivity of streamflow to future atmospheric warming and increased
snowfall. Increasing mean air temperatures resulted in a non-linear increase in total
runoff in Taylor Valley by approximately 39%, 90% and 240% for an increase of 0.5,
1 and 2°C, respectively. In addition, increasing air temperatures by only 1°C ensured
the sensible heat flux was positive (towards the surface) over the glaciers in Taylor
Valley. Increased snowfall led to decreased streamflow in cases where the snow was
enough to raise the albedo and trigger a cooling regime. However, in other cases,
snowfall contributed to increased streamflow through a greater mass turnover. This
is due to the relatively warmer and wetter air masses. These air masses decrease the
gradient between the surface and atmosphere, reduce the turbulent heat fluxes, and
result in more energy being available for melt. The research findings of this thesis
demonstrate the sensitivity of hydrological connectivity in the MDV to atmospheric
processes and the urgent need for regional-scale climate projections for this region
on the cusp of change.