Most water balance studies in the High Arctic indicate that the weather stations underestimate annual precipitation, but the magnitude of such error is unknown. Based on up to seven years of field measurements, this study provides a comparison of snowfall at weather stations with the winter snow accumulation in their nearby drainage basins.Snowfall is the major form of precipitation in the polar region for nine months every year. Without vegetation, snowdrift is controlled by the local terrain. By establishing the snow characteristics for different terrain types, total basin snow storage can be obtained by areally weighting the snow cover for various terrain units in the basin. Such a method was successfully employed to compute total winter snowfall in the drainage basins near Resolute, Eureka and Mould Bay. Results show that the basins had 130 to 300per cent more snow than the weather stations recorded. Using revised snowfall values that are reinforced by Koerner's snow core measurements from ice-caps, it is hoped that a more realistic precipitation map can be provided for the High Arctic.RÉSUMÉ La plupart des études entreprises sur le bilan hydrologique de l'arctique ont révélé que les stations météorologiques sousestiment la précipitation annuelle, la grandeur de cette erreur n'est cependant pas connu. On effectue dans la présente étude, à partir de données recueillies au cours de sept années, une comparaison entre les hauteurs de neige prélevées aux stations et celles accumulées dans les bassins de drainage à proximité de ces stations.Dans la région polaire, la précipitation principale est sous forme de neige et ceci au cours de neuf mois de l'année. En l'absence de végétation, la distribution de la congère est dictée par le relief local. En établissant les caractéristiques de la neige pour différents types de terrain, on a pu calculer l'accumulation de la neige totale du bassin en pondérant la surface de la couverture nivale par les diverses unités de terrain du bassin. Cette méthode a été appliquée avec succès pour le calcul de la hauteur totale de la neige au cours de l'hiver dans les bassins de drainage de Resolute, Eureka et Mould Bay. Les résultats indiquent que les bassins reçoivent des hauteurs de neige qui excèdent de 130 à300pourcent ceux rapportées par les stations météorologiques. En utilisant des valeurs corrigées des chutes de neige, confirmées par les mesures de Koerner à partir des carottes de neige des calottes glaciaires, on espère dresser des cartes plus réalistes de la précipitation dans l'arctique. Comparison of Weather Station Snowfall With Winter Snow Accumulation / 313 1 IntroductionAccurate information on annual precipitation is essential for water balance studies, both on a local and a regional scale (Lvovitch, 1973;Walker and Lake, 1975). Dingman ( 1973) documented many studies in the Canadian Arctic that despaired at the inaccuracies of precipitation data (e.g. Hare and Hay, 1974). Unless such basic information as winter snowfall can be established reliably, any research on water ...

High arctic slopes have a shallow active layer that thaws unevenly during summer. The result is often a lack of agreement between the configuration of the frost table and the surface topography, and this has effects on the slope hydrology. (1) Areas with a shallow frost table favour surface runoff but areas with a deeper frost table require a thick zone of saturation to generate surface flow. Uneven thaw depths then cause alternating seepage and re-emergence of water down a slope. (2) The configuration of the frost table is highly dynamic, causing day to day changes in water storage capacity in the active layer. (3) A frost table with local depressions can pond up groundwater, which may be rapidly released when part of the frozen sill is breached by continual thawing. (4) The topographical drainage divide may not correspond with the subsurface drainage divide as defined by the frost table, thus allowing groundwater to drain laterally across topographical boundaries. These findings show that a knowledge of the frost table behaviour, both spatially and temporally, is essential to the study of slope hydrology in continuous permafrost terrains.

No abstract

Installations at a High Arctic experimental site that is underlain by continuous permafrost allowed the measurement of slope runoff. Surface flow was collected near the base of the slope and the water was led to a flume with a V-notch weir. The water level in the flume was recorded and subsequently converted to discharge measurements. Subsurface flow was intercepted by an impermeable flow barrier set in a trench dug down to the permafrost table and later back-filled by the excavated slope materials. Water draining from above the flow barrier was fed into another flume unit similar to that for surface runoff. During the operational period, regular inspection of the flumes was required to ensure the prevention of ice formation or evaporative losses from the water in the flumes. Keywords: surface flow, subsurface flow, slope hydrology, permafrost.

Surface flow on arctic slopes is due to a rise of the suprapermafrost water table above the ground surface. The active layer therefore has to be saturated, or surface flow will be lost to infiltration.In spring, snowmelt supplies abundant water to the slopes, but a shallow frost table restricts the active layer storage capacity. This produces extensive surface flow, delivering considerable volumes of water downslope as overland flow or rill runoff. As the snowpacks are reduced and ground thaw increases the subsurface storage capacity, surface flow declines in magnitude and in areal coverage. During most summers, subsurface flow dominates, except when occasional rainstorms of high intensity raise the water table above the ground to rejuvenate surface runoff.The occurrence of surface flow is controlled by slope material, slope profile, and the water balance. Spatial and temporal changes in these controlling variables always result in a highly dynamic pattern of surface flow on arctic slopes.