95th ESA Annual Meeting (August 1 -- 6, 2010)

COS 38-5 - Pedogenic thresholds and soil process domains: Biogeochemical controls and ecological implications

Tuesday, August 3, 2010: 2:50 PM
408, David L Lawrence Convention Center
Peter M. Vitousek, Department of Biology, Stanford University, Stanford, CA, Oliver A. Chadwick, Department of Geography, University of California, Santa Barbara, CA, George E. Hilley, Geological and Environmental Sciences, Stanford University, Stanford, CA and Thegn N. Ladefoged, Department of Anthropology, Auckland University, Auckland, New Zealand
Background/Question/Methods

In the course of evaluating relationships between soil fertility and indigenous agriculture in Kohala, Hawaii, we collected soil samples from sites with similar parent materials across a range of mean annual precipitation spanning <200 to >3500 mm/yr, including more than 400 surface samples (integrated 0-30 cm depth) and more than 50 soil profiles (from 1 to 5 m deep).These soils were analyzed for nutrient availability (resin-extractable phosphorus, exchangeable cations) and for total pools of elements, including the immobile index elements niobium and zirconium. In earlier work with a more limited set of soils, we observed that soil properties do not change linearly in response to rainfall. Rather, their response is characterized by pedogenic thresholds associated with non-linear but predictable changes in soil properties, and by soil process domains in the regions between thresholds. Here we use our library of soils to evaluate the distribution, controls, coherence, and implications of pedogenic thresholds and soil process domains. How do weathering and leaching, the biological uplift of nutrients, the formation of secondary minerals, and oxidation-reduction dynamics interact to control element pools in surface soils? How do those element pools in turn control the biological availability of essential nutrients?

Results/Conclusions

Kohala soils sampled in upland slope positions fell into four major process domains, divided by three thresholds. At rainfall <~500 mm/yr, pedogenic carbonate influenced soil structure and chemistry, and phosphorus availability was moderate (50-100 mg/kg). From 500 - 1700 mm/yr, base saturation ranged from 30-80%, and phosphorus availability was high (100-300 mg/kg). Biological uplift enriched surface soils in phosphorus (to twice parent material levels) and iron (marginally).  From 1700 - 3000 mm/yr, primary minerals were depleted by weathering, and soils were acid (< pH 4.5) and infertile; aluminum had been mobilized from surface soils to a greater extent than iron, and atmospheric deposition was the main source of essential cations. Above ~3000 mm/yr, iron reduction has led to a greater mobility of iron than aluminum, leading to a greater mobilization of phosphorus, and carbon:nitrogen ratios increased substantially. Both indigenous Hawaiian rain-fed agriculture and modern cattle ranching depend on the second-driest process domain. These process domains also occur on substrates that are much older and younger than Kohala (150 – 400 kyr), but some of the thresholds between them occur at progressively lower rainfall on progressively older substrates.