GLACIAL MELTWATER MODELING TO SIMULATE STREAMFLOW AND LAKE LEVELS IN TAYLOR VALLEY, ANTARCTICA

The McMurdo Dry Valleys (MDV) are the largest ice-free region (4,500 km²) in Antarctica. The MDV are a polar desert with an average annual temperature of -18˚C and minimal precipitation, < 50 mm w.e. a^-1. In Taylor Valley (77°35’ S, 163°00’ E), a closed-basin, perennially ice-covered lakes occupy the valley floor. Ephemeral streams transfer glacier meltwater for ~10 weeks each summer. Glacial meltwater accounts for nearly the total inflow to these streams and lakes, groundwater is essentially non-existent. A microbially-dominated ecosystem in Taylor Valley depends on glacier runoff and thus is highly sensitive to changes to the hydroclimatic regime. A model of water supply in Taylor Valley will aid in predicting ecosystem response to a changing climate.

Mean summer air temperatures are below 0˚C so glacier ablation shows a complex sensitivity to solar radiation and wind speed, elevating the need for a distributed, physically-based energy balance model tuned specifically to local conditions. The ICEMELT model (Hoffman et al., 2016), driven by local weather measurements and calibrated using glacier ablation measurements, is applied to simulate streamflow and lake level from 1995 to 2015. The model accounts for solar radiation penetration into the ice, the spatial variability of albedo, and glacier topography that affects microclimate. Meltwater inflow, sublimation from the lake surface, and basin geometry, were used to calculate lake level.

Initial model results show that initiation and peak timing of streamflow are modeled well. Seasonal and daily flow volume correspond well to measured values for streams sourced from two glaciers, but are under-predicted for streams flowing from two other glaciers. Lake level predictions are good for those lakes that drain the well-modeled glacial streams. Reasons for the poor performance of the other two glaciers/streams are unknown. Overall, the model performed well in predicting the response to an anomalous warm event during the austral summer of 2001-02. A subsequent shift to rising lake levels in the years following the warm event, despite modest changes in air temperatures are also modeled well and will be discussed.