Comparison of Landsat TM-derived and ground-based albedos of Haut Glacier d'Arolla, Switzerland

Measurements of the surface albedo of Haut Glacier d'Arolla (Switzerland) d erived from Landsat Thematic Mapper data (alpha(TM)) are compared with grou nd-based albedo measurements (alpha(g)). The analysis is based on (a) 137 g round-based measurements made at 67 sites during three extensive glacier su rveys in the 1993 ablation season, and (b) data acquired during three corre sponding cloud-free Landsat overpasses on 29 May, 1 and 17 August. For the first overpass large differences in the snow albedo are found (average: alp ha(TM)= 0.73, alpha(g)= 0.55), which are attributed to snowfall and metamor phism of snow between the times of the satellite overpass and the ground me asurements. For the second overpass there is much better agreement between the two snow albedos (average: a,,= 0.56, a,= 0.51). For the second and thi rd overpass there is good agreement between a,, and a, of glacier ice (aver age: alpha(TM) = 0.17, alpha(g)= 0.19). However, the scatter in alpha(TM)-a lpha(g) is large which is very probably related to the high spatial variabi lity of the ice albedo over distances of only a few metres and the small-sc ale coverage tin terms of ground area of the ground measurements compared t o the TM pixel size. In addition, there is a tendency for a, to overestimat e low values of a, and to underestimate high values. This is explained by a possible bias in the ground-based measurements towards selecting either re latively clean or relatively dirty ice surfaces. Several errors associated with uncertainties in the method of albedo retrieval are discussed. It is s uggested that the assumption of an isotropically reflecting surface is a ma in source of error in the satellite-derived albedo.