Ablation of snow over sea ice is an important physical process affecting the Arctic surface energy balance. An improved understanding of the spatial and temporal variations in snowmelt onset could be utilized to improve climate simulations in the Arctic, as well as monitor the Arctic for signs of climate change. Utilizing an updated approach for monitoring snowmelt onset over Arctic sea ice, spatial variability in passive microwave derived snowmelt onset dates is examined from 1979 through 1998. The improved technique, termed the advanced horizontal range algorithm (AHRA), utilizes temporal variations in 18/19 GHz and 37 GHz passive microwave horizontal brightness temperatures obtained from the scanning multichannel microwave radiometer (SMMR) and the Special Sensor Microwave/Imager (SSM/I) to identify snowmelt onset. A qualitative assessment of spatial variability in snowmelt onset discusses the 1979 through 1998 mean snowmelt onset pattern, and it also illustrates that there are significant variations in snowmelt onset on an annual basis. Principal component analysis of the snowmelt onset dates suggests snowmelt onset variability is dominated by a zone of abnormally early (late) snowmelt onset near the Siberian coast and another zone of abnormally late (early) snowmelt onset near Baffin Bay. Statistical analysis between the first principal component and March-June monthly averaged Arctic Oscillation values implies that variations in snowmelt onset are related to alterations in the phase of the spring Arctic Oscillation. ¿ 2001 American Geophysical Union