Landscape Age as a Major Control on the Geography of Soil Weathering

Abstract Global‐scale models of rock‐derived nutrient availability often assume that physical erosion drives soils toward an approximate “steady state” over geologic timescales. By definition, steady‐state models do not represent landscape age—that is, the time elapsed since soil formation is initiated by major erosional or depositional events. We hypothesize that this steady‐state assumption has large consequences on estimates of soil fertility because landscape age can mediate the retention of mobile elements in soil, particularly in low‐relief landscapes and humid climates. We quantified the effect of landscape age on soil fertility by estimating Na retention in soils across the United States and explicitly resolving landscape age in regions that experienced significant deposition or glacial retreat after the Last Glacial Maximum (LGM). We then used a simple one‐compartment model to simulate soil formation and weathering, comparing predictions that incorporated landscape age with those based on the steady‐state assumption. We found that soils formed in LGM deposits in low‐relief, humid settings generally retain 10 times more Na than soils formed outside of LGM deposits. Furthermore, the model that accounted for landscape age outperformed a steady‐state model across the United States and increased globally averaged estimates of Na retention by 17%. These results reinforce the idea that landscape age is a major control on weathering and should not be ignored in simulations of nutrient cycling.