Which Soil Carbon Fraction is the Best for Assessing Management Differences? A Statistical Power Perspective

• Published in: Soil Science Society of America journal Vol. 79; no. 3; pp. 848 - 857
• Main Authors: Ladoni, Moslem, Basir, Abdul, Kravchenko, Alexandra
• Format: Journal Article
• Language: English
• Published: The Soil Science Society of America, Inc 01.05.2015
• ISSN: 0361-5995; 1435-0661
• DOI: 10.2136/sssaj2014.10.0426

Active fractions of soil C such as particulate organic C (POC) and short‐term mineralizable C (SMC) respond faster than total organic C (TOC) to management induced changes in soil C. However, the active fractions of organic C can possibly have larger variability that decreases the detectability of management effects on soil C. The objectives of this study were to (i) assess the relative usefulness of TOC, POC, and SMC as criteria of management induced changes on soil C and (ii) investigate if using auxiliary soil and topographical information can aid in increasing the usefulness of these criteria in studies conducted across large spatial scales. Data were collected at locations with two contrasting topographical positions (slope and depression) within 10 agricultural fields in conventional and cover crop based row crop managements at the 0‐ to 20‐, 35‐ to 50‐, and 70‐ to 90‐cm depths. The results showed that to detect differences between the management systems with an acceptable type II error of 0.20, an 80% difference in TOC and a 50% difference in SMC were needed. The statistical power for POC was never in an acceptable range. The use of auxiliary soil and topography information via analysis of covariance decreased the sizes of the minimal detectable differences. Given the faster reaction to management of SMC as compared with TOC, and its lower variability as compared with POC, we recommend SMC as the preferred C fraction for detecting treatment induced differences in organic C stocks in agricultural field experiments, especially in deeper soil layers.