The current high interest in the magnetism of loess-paleosol
sequences was sparked during the past decade by examples of a
close correspondence between fine-scale magnetic-susceptibility
variations in the Chinese loess sections and the deep-sea oxygen
isotope record [ Heller and Liu, 1984; Kukla et al.,
1988, 1990]. Magnetic susceptibility, the ratio of induced
magnetization to the inducing magnetic field, is directly
proportional to the quantity of strongly magnetic minerals in a
sample. This correspondence showed clearly that these windblown
(eolian) deposits potentially contain a high-resolution record of
climate change. Relatively high magnetic susceptibility
characterizes the paleosols that formed under relatively warm,
humid interglacial periods, and it correlates with negative
O values in deep-sea sediments. In contrast,
relatively low magnetic susceptibility characterizes loess
deposited under relatively cold, dry glacial periods, and it
correlates with positive 
O values in deep-sea
sediments. (
O is the per mil deviation of the
relative proportion of the isotopes 
O and 
O in a
sample from that in a water standard, the Standard Mean Ocean
Water.) The glacial-interglacial periods are also reflected
in the loess-paleosol sequences by different paleontologic,
geochemical, and mineralogic characteristics, including gastropod
and micromammalian fauna, Fe
/Fe
ratios, calcium
carbonate contents, clay composition, illite crystallinity,
and bulk sediment grain size.
In the broader view of atmospheric circulation, the loess and paleosol layers of the Chinese loess plateau reflect dominant monsoonal patterns [e.g., Zhongli et al., 1992; Zhang et al., 1994]. Dust was deposited on the loess plateau during glacial periods when northern hemisphere insolation was low and the summer monsoon on Asia was diminished, resulting in frequent dust falls during spring and summer. Such conditions are analogous to those of current winter months that bring northwest winds related to the Mongolian anticyclone. Paleosols formed during warmer and more humid interglacial periods associated with the southerly circulation that causes the summer monsoon, resulting in much lower dust-fall rates. The loess-paleosol sequences thus have been linked to changes in the general circulation for the northern hemisphere.
Magnetics results published before the current quadrennium
generated some essential understanding about the age and
stratigraphic framework of the loess sequences, established
testable hypotheses about the causes of enhanced magnetic
susceptibility, and helped define some outstanding problems.
These contributions engendered the explosion of recent studies.
First, paleomagnetic stratigraphy established time lines and
determined the base of Chinese loess to be about 2.5 Ma (2.5 x
10
years) old [e.g., Liu et al., 1985; Heller and
Liu, 1986; Kukla and An, 1989]. Second, ultrafine
magnetite/maghemite was recognized as an important source of the
magnetic susceptibility signal [ Kukla et al., 1988]. In
addition, the relative concentrations of magnetic particles in
the paleosols that are responsible for the high magnetic
susceptibility were attributed primarily to either (1) dilution
of a constant fallout of detrital magnetic particles by
deposition of loess having relatively low magnetic-particle
content [ Kukla et al., 1988], or (2) in-situ alteration of
paleosols by sediment compaction and leaching of carbonate [
Heller and Liu, 1984]. Zhou et al. [1990] recognized that
at least some of the magnetic signal in paleosols originated from
magnetic phases produced by soil-forming processes (pedogenesis).
Finally, the magnetic susceptibility variations were dated
independent of astronomical chronology and were interpreted to
contain orbital frequencies [ Kukla et al., 1990; Wang
et al., 1990].