Agriculture, Ecosystems and Environment 134 (2009) 251–256

Contents lists available at ScienceDirect

Agriculture, Ecosystems and Environment journal homepage: www.elsevier.com/locate/agee

Determining soil carbon stock changes: Simple bulk density corrections fail Juhwan Lee a,*, Jan W. Hopmans b, Dennis E. Rolston b, Sara G. Baer c, Johan Six a a

Department of Plant Sciences, University of California, One Shields Ave., Davis, CA 95616, USA Department of Land, Air, and Water Resources, University of California, Davis, CA 95616, USA c Department of Plant Biology, Center for Ecology, Southern Illinois University, Carbondale, IL 62901, USA b

A R T I C L E I N F O

A B S T R A C T

Article history: Received 7 April 2009 Received in revised form 21 July 2009 Accepted 29 July 2009 Available online 25 August 2009

Several methods are used to correct total soil carbon data in response to land use or management changes inherently coupled with concomitant alteration to bulk density (BD). However, a rigorous evaluation of correction methods has not been conducted. We compared original, maximum, and minimum equivalent soil mass (ESM) corrections to the fixed depth (FD) method and direct C concentrations. In a simulation exercise of a tillage event that decreased BD without change in total C concentration to a depth of 0.3 m, the original and maximum ESM methods estimated changes in total C storage of 0.34 to 0.54 Mg C ha1, well within the range of field soil C variability. In contrast, the minimum ESM method estimated changes ranging from 1.19 to 1.01 Mg C ha1. In a field experiment on reduced and intensive tillage, soil C changes (0–0.18 m) were measured from May to August 2006. The maximum ESM method generally overestimated soil C changes by 0.16 to 0.60 Mg C ha1 and the minimum ESM method underestimated them by 2.67 to 0.23 Mg C ha1 compared to the original ESM method. Field-scale soil C changes (0–0.15 m) were also measured from August 2003 to June 2005 and decreased by an unrealistic 6.64 Mg C ha1 over the first 6 months after tillage when the FD method was used. In contrast, the effect of tillage on soil C could be reasonably estimated by directly comparing changes in C concentration. In a compacted agricultural soil, we found more errors in simulated C differences when using the maximum than the minimum ESM method. Regardless of the direction of BD changes, the minimum ESM method was a better choice than the maximum ESM method in native and restored grassland systems where soil C concentrations decreased through the soil profile. We conclude that (1) the FD method is often not suitable and might be less accurate than direct C concentration measurements, and (2) the maximum/minimum ESM method can be accurate depending on the conditions (e.g., increasing or decreasing BD, systems conversion type), but (3) that the original ESM method is optimal for detecting soil C changes due to land use changes or management effects. ß 2009 Elsevier B.V. All rights reserved.

Keywords: Soil carbon stocks Soil organic matter Bulk density Equivalent soil mass Tillage Land use change

1. Introduction To quantify soil carbon changes in response to land use or management, soil samples are typically collected over time and analyzed for soil C concentration. Soil C concentration is most often converted to C mass per unit area by multiplying it with bulk density (BD) to a fixed soil depth. However, soil BD can vary spatially and temporally (Amador et al., 2000; Kulmatiski and Beard, 2004). Thus, quantification of soil C stocks can be biased if derived from dried samples obtained from a fixed depth (Ellert and Bettany, 1995). Onstad et al. (1984) showed that soil BD following conventional tillage may decrease by 10% or more for a variety of soil types. Fluctuations in BD unrelated to land use change, but as a result of wetting and drying can be substantial for soils with vertic properties. Thus, disregarding BD can over- or underestimate

* Corresponding author. E-mail address: [email protected] (J. Lee). 0167-8809/$ – see front matter ß 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.agee.2009.07.006

changes in soil C in response to land use or management and provide erroneous information for soil C sink and source assessments (Post et al., 2001; Conant et al., 2003; Vandenbygaart and Angers, 2006; Wuest, 2009). Although these effects are intuitively clear and can be observed in the field, few studies have demonstrated and quantified their influence on soil C storage estimates. Ellert and Bettany (1995) demonstrated that the equivalent soil mass (ESM) correction should be used when comparing soil C stocks in genetic horizons among land use or management practices. Equivalent soil mass is defined as the reference soil mass per unit area chosen in a layer and equivalent C mass is C mass stored in an ESM (Ellert et al., 2001). Numerous studies recognize the importance of comparing soil C stocks on the same soil mass per unit area (Ellert et al., 2001; Gifford and Roderick, 2003; Vandenbygaart, 2006; Vandenbygaart and Angers, 2006). The equivalent C mass calculation is expected to reduce sampling errors in estimates of soil C due to differences in the amount and placement of plant material input throughout soil profiles under

J. Lee et al. / Agriculture, Ecosystems and Environment 134 (2009) 251–256

252

various management activities. For example, soil C concentrations can be enriched or diluted by differences in sampling depth, where a concentration gradient of C with soil depth is evident (Angers et al., 1995; Staricka et al., 1991). This suggests that BD influences the soil C stock estimated across soil profiles, particularly at the soil surface where the majority of the roots and residue exists. Nevertheless, soil mass corrections have rarely been conducted, and there is still a lack of evaluation of the multiple approaches to correct for soil mass. Therefore, a rigorous evaluation of how to select which soil mass to be used as the ESM is needed. In this study, we present multiple ESM methods alternative fixed depth (FD) corrections used to calculate changes in soil C using simple simulations and field data. The field situations were representative of a wide range of agricultural/ecological research objectives: short-term (