Using Magnetic Remanence Parameters to Reconstruct Past Precipitation From Buried Soils.

Magnetic measurements that rely on the comparisons between magnetically enhanced paleosols and their unaltered parent materials have long been used to quantify magnetic enhancement and to reconstruct past climatic conditions. We present measurements of magnetic remanence parameters (ARM/IRM, magnetic coercivity distributions) which, for many sites, allow for the direct quantification of pedogenically produced magnetite and estimates of past climatic conditions. Since our method does not rely on comparisons between magnetically enhanced soil horizons and their presumably unaltered parent material, this method might be especially suited for the analysis of thin loess-paleosol sequences, where pedogenically unaltered loess is absent. To test our method we analyzed the Brady soil in several locations in eastern Colorado, southwestern Nebraska and northwestern Kansas. Our analyses yielded mean annual precipitation values of 450 mm/a (Wauneta, 40.4983 ° N, -101.404 ° W), 550 mm/a (Devil's Den, 41.4561 ° N, -100.192 ° W) and 575 mm/a (Harlan County Lake, 40.0709 ° N, -99.2737 ° W), which is slightly lower than present-day precipitation at these sites and correlates well with isotope-derived reconstructions of paleoprecipitation. These quantitative reconstructions should be considered minimum estimates of paleoprecipitation as the method assumes a relatively rapid formation of the magnetic signal and no significant alteration of the magnetic phase after burial.