Atmospheric conditions and fast ice at Davis, East Antarctica: A case study

We present in situ observations of atmospheric and fast‐ice parameters at Davis (68°35′S, 77°58′E), East Antarctica. Surface measurements from 1969–2003 reveal negative trends for summer and autumn surface‐air temperatures, yet positive annual, winter and spring trends. Mean sea level pressure decreased over the observational record both annually (−0.10 hPa yr −1 , significant at the 90% confidence level (90)) and seasonally. We mainly associate this with increased cyclonic activity. These trends reflect large‐scale circulation changes identified by others in the troposphere and lower stratosphere. Temporal changes in the Davis landfast‐ice characteristics are evaluated using in situ measurements from the late 1950s together with intermittent data from 1979 onward. There is large interannual variability in fast‐ice characteristics, which peaked during the 1990s. Mean values of annual maximum ice thickness taken over subintervals have changed little. However, dates of annual maximum ice thickness and final fast‐ice breakout are delayed (each by +0.43 d yr −1 ). Delays in ice breakout contribute to a prolonged persistence of the fast ice (+0.67 d yr −1 , (90)). Linear regression was applied to relate changes in fast‐ice characteristics to those in the atmosphere. Significant at the 90% level or above are: Winter temperature and annual maximum ice thickness are anticorrelated ( R 2 = −0.69); the tendency toward later dates of annual maximum ice thickness appears influenced by winter ( R 2 = +0.47) and spring ( R 2 = +0.42) warming. Interannual changes in spectral density of the (annual and summer, respectively) mean sea level pressure within the cyclone‐active frequency band are crucial to annual maximum ice thickness ( R 2 = −0.54) and fast‐ice breakout ( R 2 = −0.59). Maximum ice thickness is furthermore influenced by winter wind speed ( R 2 = −0.52).