A Long‐Term Field Bioassay of Soil Quality Indicators in a Semiarid Environment

A major limitation of traditional approaches to quantifying the relationship between soil quality and productivity is the confounding effect of landscape quality factors such as topography, hydrology, and climatic parameters. In this 14‐yr study, we used a soil‐transplant field bioassay under uniform landscape conditions to identify key soil quality attributes that could be related to spring wheat ( Triticum aestivum L.) biomass production in southern Alberta, Canada. Thirty‐six soils (main plots) collected from sites with divergent cropping and management histories were deposited at a common site from which topsoil had been removed. Total biomass yield, assessed with or without N fertilizer application (subplot), was used as an integrator of 24 soil quality indicators tested. Although between‐soil variability in biomass production differed significantly among the years, we found no evidence of productivity convergence among the 36 soils after 14 yr. Partial least squares analysis identified total organic C (TOC), total inorganic C, total N (TN), light fraction (LF)‐C, LF‐N, mineralizable C and N (C min and N min ), and extractable nutrients (N and P) among the most important soil quality indicators associated with variation in biomass production. Critical concentrations, above which no significant yield response to additional indicator level was expected, were estimated using segmented‐model regression analysis. Nitrogen application increased critical concentrations for LF‐C and LF‐N and decreased those for N min and Olsen P, but had no significant effect on TOC and TN critical concentrations and suppressed yield response to C min and KCl‐extractable N concentrations. Our findings provide evidence that selected indicators can provide a definitive, quantitative assessment of soil quality and lend credence to the value of our field approach in quantifying relationships between soil function and indicators for specific areas.