398-4 Using Buried Soils to Reconstruct Past Climate: A Quantitative Approach.

See more from this Division: S05 Pedology
See more from this Session: Soil Genesis and Classification: II
Wednesday, October 24, 2012: 1:50 PM
Duke Energy Convention Center, Room 235, Level 2
Share |

Ashley B. Zung1, Johannes J. Feddema1 and Rolfe D. Mandel2, (1)Geography, University of Kansas, Lawrence, KS
(2)Kansas Geological Survey, University of Kansas, Lawrence, KS
Climatic factors that govern soil water supply and water movement through the soil strongly influence pedogenesis. Pedogenic pathways produce predictable soil properties, and A horizon organic carbon content, calcium carbonate accumulation in the subsoil, and clay translocation through the soil solum of modern soils have been quantitatively linked to modern climate (e.g. Rasmussen et al. 2005; Dai and Huang 2006; Gray et al. 2009; Scull 2010). Climate is arguably the dominant factor dictating pedogenic pathways on the Great Plains, U.S., where topographic and geologic spatial gradients are quite subtle. Buried soils, representing past episodes of landscape stability and soil formation, are commonly preserved in alluvial and eolian deposits on the Great Plains. Because soils are often preserved in the stratigraphic record and climate significantly effects pedogenesis in the region there is potential to quantitatively reconstruct past climate on the Great Plains based on buried soil properties.

We conducted a statistical analysis of modern soil-climate relationships on the Great Plains as a proxy for soil-climate associations during the late Quaternary. Soil characterization data from the NCSS Soil Characterization database and long-term climatologies from the University of Delaware Center for Climatic Research Climate Data Archive were used to develop multiple regression models. Soils that served as modern analogues to those preserved in alluvial and eolian sequences on the Great Plains were selected from the database based on soil order, parent material, and geography. Models predicting mean annual precipitation and annual moisture index based on soil properties were highly significant (R2 > .3, p < .0001), with residual standard errors of 117 mm and 0.14, respectively. Efforts to control for potential confounding factors, such as biotic and geomorphic factors and time, and considerations for applying the model in the field will also be discussed.

See more from this Division: S05 Pedology
See more from this Session: Soil Genesis and Classification: II