Modelling leaf wax n-alkane inputs to soils along a latitudinal transect across Australia

Abstract

Leaf wax n-alkanes provide a valuable palaeoecological proxy, but their interpretation requires an understanding of the scale of temporal and spatial integration in soils. Leaf wax n-alkanes are continually deposited into soils directly from local plants as well as from more distant plants via wind or water transport. In addition, n-alkanes can persist in soils for thousands of years, and tend to decrease in age with shallower depth. To explore whether the uppermost soils reflect recent leaf fall inputs we compared surface soils and modern vegetation from 20 sites along a transect across Australia. At each site, the three most dominant plant species and a soil sample from the top 3 cm were analysed for n-alkane concentration, average chain length (ACL), proportional abundance of C33 and C29 (Norm33) and carbon preference index (CPI). Chain length distributions differ between trees and grasses, with a higher proportion of C29 in trees and C33 in grasses. Norm33 in soils correlates with proportional grass to tree cover across the transect. To model n-alkane inputs for each site, we calculated a predicted ACL, Norm33 and CPI using the dominant plants at that site, weighted by proportional species cover and n-alkane concentration. Predicted ACL, Norm33 and CPI inputs were generally higher than the soils, demonstrating that recent and local inputs do not dominate soil n-alkanes at our study sites. Thus, n-alkane distributions in surface soils do not correlate with local, current vegetation, but do correlate with proportional grass and tree cover, suggesting they provide a faithful record of large scale ecosystem structure.