Errors in daily ablation measurements in northern Greenland, 1993-94, and their implications for glacier climate studies

Ablation -climate studies were made at two locations in northern Green­ land in the summers of 1993 and 1994, respec tively. Daily ablation II'as measured at ten stakes within a small area, and the data were compared with each other to detect gross errors. For example, high deviations for data taken on the same day, or low cor­ rela tions between data series at differ ent stakes, indicate erroneous data. A fter cl iscard i ng data for one stake in 1993 and t,wo stakes in 1994, random errors in daily ablation data for individ ual stakes arc ± 5 kg m 2 d I, which is further reduced to only about ± 2 kg m -2 d 1 by averag ing over eight or nine stakes. Random errors in cal culated energy bala nces using the pres ent abla tion data are much ImoVer than found in earlier studies in \ Vest Greenland where ablation was only measured on three stakes without any attem pt to detect gross errors. Aside from day-to -day errors, there arc ± 10% diffe rences in mca n ablation at different sta kes, which are probably caused by small-s cal e variations in surface albedo. Such inters take diffe rences gi"e � ± 10% uncertainty in positive degree-day faetors, which arc 9.8 ± 0.9 and 5.9 ± 0.6 kg m 2 d 1 deg 1 for the two sites.