No abstract

[1] North Atlantic variability in general, and the North Atlantic Oscillation (NAO) in particular, is a long-studied, very important but still not well-understood problem in climatology. The recent trend to a higher wintertime NAO index was accompanied by an additional increase in the Azores High not coupled to changes in the Icelandic Low, as shown by a self-organizing maps (SOMs) analysis of monthly mean DJF mean sea level pressure data from 1957 to 2002. SOMs are a nonlinear tool to optimally extract a user-specified number of patterns or icons from an input data set and to uniquely relate any input data field to an icon, allowing analyses of occurrence frequencies and transitions complementary to principal component analysis (PCA). SOMs analysis of ERA-40 data finds a North Atlantic ''monopole'' roughly colocated with the mean position of the Azores High, as well as the well-known NAO dipole involving the Icelandic Low and the subtropical high. Little trend is shown in December, but the Azores High increased along with the NAO in January and February over the study interval, with implications for storminess in northwestern Europe. In short, our SOM-based analyses of winter MSLP have both confirmed prior knowledge and expanded it through the relative ease of use and power with nonlinear systems of the SOM-based approach to climatological analysis.

[1] A computationally cheap method is described for estimating monthly means of climatological variables on alternate calendars, given their monthly means on an original calendar. This can be used to convert archived paleoclimatic GCM monthly mean results originally saved on the modern calendar to a more appropriate equi-angular calendar, thus avoiding spurious orbital signals when comparing monthly means for modern and past simulations. The method is tested using GCM simulations for the present and 126 ka BP. The method works well for temperature, which has relatively smooth seasonal cycles, and satisfactorily for other climatological variables in most regions except mid to high southern latitudes. It is also satisfactory for precipitation in most regions, but in the tropical ITCZ, it fails to significantly reduce the spurious calendar errors. However, this is largely due to high-frequency natural interannual variability in the 10-year simulations, and the errors are of the same order as the uncertainty in monthly means on a single calendar.

Ice cores have, in recent decades, produced a wealth of palaeoclimatic insights over widely ranging temporal and spatial scales. Nonetheless, interpretation of ice-core-based climate proxies is still problematic due to a variety of issues unrelated to the quality of the ice-core data. Instead, many of these problems are related to our poor understanding of key transfer functions that link the atmosphere to the ice. This study uses two tools from the field of artificial neural networks (ANNs) to investigate the relationship between the atmosphere and surface records of climate in West Antarctica. The first, self-organizing maps (SOMs), provides an unsupervised classification of variables from the midtroposphere (700 hPa temperature, geopotential height and specific humidity) into groups of similar synoptic patterns. An SOM-based climatology at annual resolution (to match ice-core data) has been developed for the period 1979-93 based on the European Centre for Medium-Range Weather Forecasts (ECMWF) 15-year reanalysis (ERA-15) dataset. This analysis produced a robust mapping of years to annual-average synoptic conditions as generalized atmospheric patterns or states. Feed-forward ANNs, our second ANN-based tool, were then used to upscale from surface data to the SOM-based classifications, thereby relating the surface sampling of the atmosphere to the large-scale circulation of the mid-troposphere. Two recorders of surface climate were used in this step: automatic weather stations (AWSs) and ice cores. Six AWS sites provided 15 years of near-surface temperature and pressure data. Four ice-core sites provided 40 years of annual accumulation and major ion chemistry. Although the ANN training methodology was properly designed and followed standard principles, limited training data and noise in the ice-core data reduced the effectiveness of the upscaling predictions. Despite these shortcomings, which might be expected to preclude successful analyses, we find that the combined techniques do allow ice-core reconstruction of annual-average synoptic conditions with some skill. We thus consider the ANN-based approach to upscaling to be a useful tool, but one that would benefit from additional training data.

This study examines the biases, intermodel spread, and intermodel range of surface air temperature (SAT) across the Antarctic ice sheet and Southern Ocean in 26 structurally different climate models. Over the ocean (40°–60°S), an ensemble-mean warm bias peaks in late austral summer concurrently with the peak in the intermodel range of SAT. This warm bias lags a spring–summer positive bias in net surface radiation due to weak shortwave cloud forcing and is gradually reduced during autumn and winter. For the ice sheet, inconsistencies among reanalyses and observational datasets give low confidence in the ensemble-mean bias of SAT, but a small summer warm bias is suggested in comparison with nonreanalysis SAT data. The ensemble mean hides a large intermodel range of SAT, which peaks during the summer insolation maximum. In summer on the ice sheet, the SAT intermodel spread is largely associated with the surface albedo. In winter, models universally exhibit a too-strong deficit in net surface radiation related to the downward longwave radiation, implying that the lower atmosphere is too stable. This radiation deficit is balanced by the transfer of sensible heat toward the surface (which largely explains the intermodel spread in SAT) and by a subsurface heat flux. The winter bias in downward longwave radiation is due to the longwave cloud radiative effect, which the ensemble mean underestimates by a factor of 2. The implications of these results for improving climate simulations over Antarctica and the Southern Ocean are discussed.