Geoscience Reference
In-Depth Information
may remain high if losses from evaporation are very low, even when precipitation
onto the lake and its catchment is relatively low. The example of late Pleistocene Lake
George near Canberra, Australia, may be used to illustrate this paradox.
Using the mapped lower limits of periglacial solifluction deposits in the Snowy
Mountains of south-east Australia, Galloway (
1965b
) deduced that the upper limit
attained by trees (the 'tree-line' or 'timberline') during the Last Glacial Maximum had
been lowered by at least 975 m. Because the tree-line today coincides with the 10
C
isotherm for the hottest month (in this case, January), the difference between the mean
January temperature at that elevation today and 10
°
°
C is the temperature rise since the
LGM, amounting to at least 9
C. He then used the relationship between mean monthly
temperature and mean monthly evaporation measured at Canberra to estimate an LGM
annual rate of evaporation of 510 mm from Lake George near Canberra. Using the
measured ratio between winter precipitation and run-off in a nearby catchment, he
estimated that run-off into glacial Lake George would amount to about 104 mm for an
annual precipitation of 380 mm, or only slightly more than half the present long-term
mean rainfall onto the lake. Galloway concluded that glacial Lake George was up to
30 m deep, despite receiving only half the amount of rainfall received today. This
minevaporal
hypothesis has been widely debated ever since, as has the chronology of
the 30 m high lake shoreline. (It now seems more likely, from as yet unpublished new
radiometric ages, that the 30 m shoreline may date back to a wetter interval in the late
Pleistocene preceding the Last Glacial Maximum).
Galloway (
1970
;
1983
) later extended his reasoning to a re-evaluation of the LGM
climate in the south-western United States, discussed in
Section 12.3
, raising the
question of whether the glacial climate was indeed mild and wet, as some palynolo-
gists believed (Van Devender and Spaulding,
1979
), or cold and dry, the model he
considered to be most in accord with the evidence. The issue of whether the pluvial
lakes of the south-western United States reflect a wetter or a drier climate is obscured
by the problem of estimating the source and amount of run-off into the lakes, some
of which represents glacial meltwater.
In semi-arid western New South Wales, the Willandra Lakes - of which Lake
Mungo is best known - have yielded a detailed history of high and low lake levels
over the past 40 ka, based on both luminescence and
14
Cages(Bowler,
1998
;Bowler
and Price,
1998
). Although all of the
14
C ages are in the process of being fine-tuned
at present, it is clear that the LGM was a time of low lake levels and substantial influx
of desert dust (
Wustenquartz
) into the lake floor sediments, which was brought about
by regional aridity (Bowler,
1998
).
°
12.8 Glacial aridity in tropical deserts
The alternative hypothesis to glacial pluvial climates in tropical deserts is the concept
of glacial aridity. Fairbridge (
1965
;
1970
) was an early and vigorous champion of
the hypothesis of glacial aridity and by 1975 there was broad acceptance that late
