MAGNETIC ENHANCEMENT AND PALEOCLIMATE - STUDY OF A LOESS/PALAEOSOL COUPLET ACROSS THE LOESS PLATEAU OF CHINA

The origin of the Milankovitch palaeoclimatic signal so dramatically recorded in the Chinese loess remains uncertain. At least five hypotheses have been put forward to explain the well-documented enhancement of magnetic susceptibility in palaeosols which formed when conditions were warmer and wetter. We report here a magnetic study of a prominent loess-soil couplet (L4-S3) at three widely separated sites where present-day mean annual rainfall and temperature range from similar to 350 mm to similar to 700mm and from similar to 6 degrees C to similar to 13 degrees C. The properties determined are laboratory remanences (isothermal, IRM; anhysteretic, ARM; viscous, VRM, including viscous decay), room-temperature susceptibility, and thermomagnetic behaviour. At fields beyond 100 mT, the IRM spectra are all essentially identical, both in shape and in absolute magnitude, suggesting a uniform magnetic component widespread across the Plateau. Subtracting the signal of this component isolates a second component, the amount of which varies by an order of magnitude between sites. The magnetic properties of this latter material are remarkably similar to those reported for bacterial magnetite in geological sediments and modern soils. Viscous decay of VRM takes place in two distinct stages, one of which falls to zero in <10 min. We attribute magnetic enhancement-which is the source of the palaeoclimatic signal-to in situ pedogenetic formation of magnetite, which we characterize in our interpretation as two separate forms of bacterial magnetofossils. The most magnetic sample requires a mass fraction of about half of one per cent of this biogenic material