Influence of seasonality on glacier mass balance, and implications for palaeoclimate reconstructions

Abstract

Golledge, N., Hubbard, A., Bradwell, T. (2010). Influence of seasonality on glacier mass balance, and implications for palaeoclimate reconstructions. Climate Dynamics, 35(3), 757-770.Climates inferred from former glacier geometries in some areas exhibit discrepancies with regional palaeoclimates predicted by General Circulation Models (GCMs) and modelling of palaeoecological data, possibly as a consequence of their differing treatments of climatic seasonality. Since glacier-based climate reconstructions potentially offer an important tool in the calibration of GCMs, which themselves need validation if used to predict future climate scenarios, we attempt to resolve mismatches between these techniques by (1) investigating the influence of seasonality on glacier mass balance, and (2) refining the methodology used for the derivation of glacier-based palaeoclimates. Focussing on the Younger Dryas stadial glaciation of Scotland, northeast Atlantic, we show that sea-ice amplified seasonality led to a significantly drier climate than has been suggested by glacier-based interpretations. This was characterised by a relatively short ablation season and the survival of a more substantial winter snowpack. We suggest that if palaeoglaciological studies were to account for changes in seasonal temperature and precipitation variability, their results would agree more closely with the cold, arid, northeast Atlantic palaeoenvironment predicted by atmospheric modelling and northwest European pollen studies, and would therefore provide more accurate constraints for GCM calibration.Peer reviewe

Evolution of a Lateglacial mountain icecap in northern Scotland

“…The alternative function allows for an annual temperature range (mean July T -mean Jan T = 30°C) that was three times greater than today, and so is more aligned with evidence from some palaeoclimatic proxies (e.g. Atkinson et al 1987;Isarin et al 1998;Witte et al 1998;Lie & Paasche 2006): (2) is a scaling coefficient allowing for seasonality in precipitation ( = 1 for neutral precipitation seasonality where daily precipitation = 1/365 x total annual precipitation; = 1.4 for summer dominated precipitation; and = 0.8 for winter dominated precipitation; Golledge et al 2010). It is also useful to consider a function that assumes a modern maritime annual temperature range (mean July T -mean Jan T = 10°C; Golledge 2008), which is derived from the original modelling experiments presented by :…”

“…Seasonality was considered to follow present (maritime) annual temperature ranges, and modern precipitation values were invoked over western Scotland, but with 60% northward and 80% eastward reductions (see for more discussion). In subsequent experiments, Golledge et al (2010) and Golledge (2010b) found that similar modelled glacier configurations across Scotland could equally be produced with doubled or trebled seasonality, combined with reduced total precipitation, and more relaxed precipitation gradients; thus demonstrating that different climatic regimes can produce similar overall ice mass dimensions, albeit with different glaciological conditions. For the purposes of this discussion, comparisons are made with the simulation of Golledge et al ( , 2009, which is most fully described.…”

“…is a scaling coefficient allowing for seasonality in precipitation ( = 1 for neutral precipitation seasonality where daily precipitation = 1/365 x total annual precipitation; = 1.4 for summer dominated precipitation; and = 0.8 for winter dominated precipitation; Golledge et al 2010). It is also useful to consider a function that assumes a modern maritime annual temperature range (mean July T -mean Jan T = 10°C; Golledge 2008), which is derived from the original modelling experiments presented by :…”

“…However, recent studies by and Golledge et al (2010) have shown that use of such a global dataset neglects the regional influence of seasonality upon former Scottish ice masses during the YD, and may have resulted in over-estimates of palaeoprecipitation. As a result, Golledge et al (2010) used numerical model output to propose an alternative precipitation-temperature function specific to the Scottish YD environment. The alternative function allows for an annual temperature range (mean July T -mean Jan T = 30°C) that was three times greater than today, and so is more aligned with evidence from some palaeoclimatic proxies (e.g.…”

A Younger Dryas plateau icefield in the Monadhliath, Scotland, and implications for regional palaeoclimate

“…Benn and Ballantyne, 2005;Finlayson et al, 2011). Its suitability for palaeoclimatic reconstruction has been debated, however, due to differences in incoming radiation and factors such as aspect, wind direction and topography that may affect the precipitationetemperature relationship at a local scale (Dahl and Nesje, 1996;Dahl et al, 1997;Kaser and Osmaston, 2002;Evans, 2006;Braithwaite, 2008;Golledge et al, 2010). Others argue that this 'smoothing-out' of local variations is advantageous, since it makes the dataset more reliable compared to attempting to account for local variability in areas where no modern glacier data exists (Benn and Ballantyne, 2005;.…”