Abstract
The turbulent sensible and latent heat fluxes are important components of the surface energy balance over glaciers in the Southern Alps of New Zealand, contributing over half the energy available for ablation during large melt events. To calculate these terms confidently in glacier mass-balance models it is essential to use appropriate parameterizations for surface roughness and atmospheric stability. Eddy covariance measurements at Brewster Glacier were obtained over an ice surface to help facilitate an assessment of the calculation of the turbulent heat fluxes. The roughness length for momentum was found to be 3.6 x 10(-3) m, while the roughness lengths for temperature and humidity were two orders of magnitude smaller, in agreement with surface renewal theory. A Monte Carlo approach was used to assess the uncertainty in turbulent heat fluxes calculated using the bulk aerodynamic method. It was found that input-data and roughness-length uncertainty could not explain underestimates of observed sensible heat fluxes during periods with low wind speed and large temperature gradients. During these periods a katabatic wind speed maximum alters the formulation of the turbulent exchange coefficient to that typically observed in a neutral atmosphere and this has implications for glacier mass-balance sensitivity.