Energy balance and runoff seasonality of a Bolivian glacier

The runoff of Zongo Glacier (Bolivia, 16°S) shows an appreciable seasonal variability, with low discharges in the dry season (May to August) and high values in the humid season (October to March). Incoming radiation, temperature and precipitation are poorly correlated with discharge and cannot explain the hydrological seasonality and the glacierʹs response to climate variability in the Tropical Andes. Since 1996, energy balance measurements have been carried out in the vicinity of the mean equilibrium line (5150 m a.s.l.) on Zongo Glacier (2.1 km2). Comparisons are made with proglacial stream discharges recorded at the main hydrometric station. Each component of the energy balance (net radiation, turbulent heat fluxes, heat transfer into the ice and heat supplied by precipitation) is derived separately from the measurements, and the variability throughout the year is evaluated. Radiation and turbulent fluxes dominate the surface energy balance. Sensible heat flux is small and does not show a significant seasonal change. Latent heat flux is highly variable with low values during the accumulation season and high values during the dry period. This high sublimation loss during the dry season causes well-developed penitents at the glacier surface. In conclusion, incoming energy throughout the year is constant, with no large variations, and humidity controls the balance of this energy between sublimation and melting. During the accumulation season, sublimation is reduced because of a low gradient of vapour pressure and the energy supplied by radiation is directly consumed by melting, explaining why discharge is high. During the dry period, a large part of the energy supplied by radiation is used to sublimate snow or ice and therefore, energy available for melting is low, which leads to low melt rates. Due to the important role of humidity, tropical glaciers are likely to be the climatic indicators the most sensitive to climatic changes like the greenhouse effect.