Geoscience Reference
In-Depth Information
The brilliant work by Professor Liu Tungsheng and his colleagues demonstrated that
beds of unweathered loess laid down during cold, dry climatic episodes alternated
with beds of highly weathered loess in which often well-developed soils (Kemp,
2001
) had formed during warm, wet climatic interludes, as indicated by the pollen
and fossil faunal remains (especially mollusca) preserved within them (Kukla,
1987
;
Liu and Ding,
1998
; Kohfeld and Harrison,
2001b
). A simple notation system was
developed in which L refers to fresh loess beds and S to fossil soils, numbered from
the Holocene (S1) back through the LGM (L1) to the base of the sequence. Magnetic
susceptibility analyses revealed that the red paleosols contained iron formed during
pedogenesis - iron lacking in the unweathered loess beds (Kukla,
1987
;Evansand
Heller,
2001
; Maher et al.,
2010
). Particle size analysis showed that coarser grains
were more abundant during the cold, dry phases of loess deposition. Source areas
were the Taklamakan Desert and several sand deserts in the Gobi and north-west
Inner Mongolia, including the Badain Jaran, Tengger and Mu Us deserts (Ding et al.,
1999
), as well as central Asia (Sun et al.,
2010
).
The initial interpretation of the alternating loess and soil (L/S) sequence was that
the soils developed during times when the summer monsoons were at least as strong
as they are today (i.e., during interglacials and interstadials), and the loess units
developed during times when the winter monsoon was stronger and the Siberian
High Pressure system was more intense than they are today (i.e., during glacials
and stadials). Roe (
2009
) has challenged this interpretation, arguing that most of the
present-day dust outbreaks from western China occur in spring and are associated
with the passage of strong cold fronts that produce intense windstorms able to entrain
substantial amounts of dust. The same is true inMongolia, where the dust-storms occur
primarily in the spring (Middleton,
1991
). These events coincide with weakening of
the Siberian High, and this pattern of events is also likely to have occurred in the
past, suggesting that the simple dichotomy between 'winter' and 'summer' monsoon
is somewhat oversimplified. Another problem relates to differential preservation of
the loess sequence, with episodes of intense gully erosion coinciding with the onset
of interglacial phases (Porter and An,
2005
).
Detailed sampling of stratigraphic sections located on a set of east-west and north-
south transects has revealed a sequence of thirty-seven loess-soil couplets spanning the
past 2.5 Ma. Each couplet represents a cold and dry phase of rapid dust accumulation
and an ensuing wet and warm phase of weathering and soil formation. A soil is
defined as weathered loess showing at least as much pedological organisation as the
widespread early Holocene soil at the top of the loess sequence. Interpretation of
the loess-soil couplets is based on high-resolution sampling and detailed analyses of
grain size, magnetic susceptibility, organic carbon, sediment micromorphology and
mineralogy, calcium carbonate content and mollusc species. At a very general level,
the soils are thought to indicate a weaker winter monsoon and a stronger summer
monsoon. Conversely, the coarser-grained, unweathered loess with generally much
