Soil carbon lost from Mollisols of the North Central U.S.A. with 20 years of agricultural best management practices

https://doi.org/10.1016/j.agee.2012.08.011Get rights and content

Abstract

Soil organic carbon (SOC) is highly sensitive to agricultural land management, so there is a great deal of interest in managing cultivated soils to sequester atmospheric CO2. In this study we evaluated the influence of six cropping systems on SOC at the Wisconsin Integrated Cropping System Trial (WICST) over a 20-year period. Analysis of SOC on either a concentration or mass per volume of soil basis indicated a significant decline across all of the systems at WICST. While the rotationally grazed pasture system sequestered carbon (C) in the surface 15 cm these gains were offset by losses at depth. Both no-till (NT) practices and inclusion of perennial crops reduced SOC loss, but neither resulted in C sequestration in the soil profile. Results from this study demonstrate the importance of (i) comparing current and initial soil samples when evaluating SOC sequestration and (ii) evaluating SOC changes throughout the soil profile. The losses of SOC at depths below the plow layer point to either a lack of C input from roots, increased oxidative loss at these depths or both.

Highlights

► Effect of six agroecosystems of the North Central U.S.A. on soil organic carbon (SOC). ► Modern SOC data are compared to archived samples to ascertain ΔSOC. ► SOC change is analyzed on both a concentration and equivalent mass basis. ► In spite of agricultural practices or cropping system, SOC was lost over a 20-year period. ► Best management practices may not stabilize SOC in Mollisols of North Central U.S.A.

Introduction

The world's soils contain 2500 Pg of C, almost twice the combined amount of C in the atmosphere and vegetation (Batjes, 1996, Lal, 2008). The SOC pool makes up 60% of this total, 170 Pg of which are stored in the 1.7 billion hectares of agricultural cropland worldwide (Paustian et al., 2000, Lal, 2008). Hence, changes in soil C content can have a large effect on the global C budget (Bellamy et al., 2005). There is currently a great deal of interest in managing agricultural soils as C sinks to help offset rising levels of atmospheric CO2. This approach to atmospheric CO2 reduction is an attractive option because it is potentially cost effective and can be implemented using current agricultural technologies (Conant et al., 2007, Lal, 2008). Improvements in agricultural genetics and technology over the past 50 years have greatly increased the crop residue contribution to SOC and enabled widespread adoption of conservation and NT farming practices, both of which are thought to promote C sequestration or slow its loss (Buyanovsky and Wagner, 1998). According to the 2011 Inventory of U.S. Greenhouse Gas Emissions and Sinks compiled by the U.S. EPA, land converted to cropland in 2009 was cited as a source of CO2 (−1.6 Tg C yr−1), while land continuing in cropland and land converted to grasslands were both considered sinks of atmospheric CO2 (4.7 and 2.3 Tg C yr−1, respectively) (USEPA, 2011). In addition to the potential mitigation of climate change, increasing SOC levels has many other benefits including: (i) improved soil structure, (ii) reduced soil erosion, (iii) decreased non-point source pollution, (iv) increased water holding capacity, (v) improved cation exchange capacity, and (vi) increased soil fertility for food production (Lal, 2008).

Agricultural practices affect SOC mineralization and stabilization by directly altering soil moisture, temperature, aeration, pH, and nutrient availability. Cultural practices that are often cited as ways to increase SOC or mitigate its loss include conversion from conventional to NT farming (West and Post, 2002, Huggins et al., 2007), cultivation of perennial crops (Huggins et al., 1998), use of livestock and green manures (Ogle et al., 2005), increased crop rotation complexity (West and Post, 2002), and the application of fertilizer (Huggins et al., 1998, Nafziger and Dunker, 2011). West and Post (2002), in a meta-analysis of 67 long-term experiments, reported that conversion from conventional tillage (CT) to NT resulted in an estimated sequestration rate of 570 ± 140 kg C ha−1 yr−1 and that by increasing rotation complexity, an additional 200 ± 120 kg C ha−1 yr−1 could be sequestered. These results are consistent with a global meta-analysis of 167 experiments that showed SOC increased with land set aside, reduced tillage, and increased C inputs through cropping practices (Ogle et al., 2005).

Despite these positive findings, a similarly large body of literature has demonstrated the conditionality of C sequestration or found net losses under what are considered best management scenarios. In an extensive study comparing soil samples from 1978 to 2003, Bellamy et al. (2005) reported C losses across land use and land cover types in the UK. These trends were most pronounced in systems with high initial C content, but were not related to land management, suggesting a link to climate change. Working in southern Minnesota on a clay loam, Huggins et al. (2007) found that under different management practices for annual crops (including NT), C losses of 1.6–3.7 Mg C ha−1 yr−1 occurred over the course of 14 years. They concluded that annual cropping systems had limited potential to restore C levels to those of native sites, and that under the best scenario of continuous corn and NT management, stabilization of initial SOC levels would either require reducing C decomposition rates by over 50% or doubling C inputs. These conclusions were supported by a recent meta-analysis of 69 paired experiments comparing the impacts of CT and NT on SOC sequestration (Luo et al., 2010), who reported that cultivation of native systems for more than 5 years resulted in SOC losses in excess of 20 Mg C ha−1 in the surface 60 cm of the soil profile irrespective of cultural practice (CT or NT). When farming practices were switched from CT to NT they found that NT increased SOC accumulation in the surface 30 cm, but that when deeper soils were considered (>40 cm), SOC was stable.

Lack of statistical power, use of space as a surrogate for time, and incomplete accounting efforts can all lead to divergent conclusions about SOC stability (VandenBygaart and Angers, 2006, Sanderman and Baldock, 2010, Kravchenko and Robertson, 2011, Schmidt et al., 2011). Soil is inherently variable, making it difficult in some instances to detect management induced trends in SOC. While retrospective power analysis has been suggested as a means to improve the interpretation of SOC change results, particularly as they pertain to Type II statistical errors (Kravchenko and Robertson, 2011), the technique is not widely accepted and has been discouraged by many (Hoenig and Heisey, 2001, Lenth, 2001, Parker and Berman, 2003).

In many analyses of SOC dynamics, baseline soil samples are not available. In such instances unmanaged systems or control treatments are used to estimate historic SOC levels. While such studies provide important information on relative treatment differences, it is questionable what conclusions can be drawn regarding SOC sequestration or loss. Sanderman and Baldock (2010) indicated that less than 50% of the studies in major reviews of SOC stock changes have followed changes in SOC through time. They asserted that without adequate baseline SOC data, it is impossible to determine whether or not a measured difference between two treatments has resulted in sequestration of atmospheric CO2. In spite of the importance in evaluating SOC change over time, this method is also not without potential drawbacks. If for example a system is not at a state of SOC equilibrium at the time when baseline soil samples are taken, attribution of changes in SOC to current or previous land management may be difficult (Sanderman and Baldock, 2010).

While most of the changes in SOC associated with agricultural management appear to occur within surface soil horizons (Syswerda et al., 2011), ignoring SOC trends in deeper horizons and/or changes in bulk density (BD) can result in significant errors in the estimation of SOC change. Baker et al. (2007) showed that the differences between NT and CT primarily stemmed from differences in SOC distribution and not SOC accumulation in NT. In addition to whole profile analysis, corrections must often be made to account for changes in bulk density. Such corrections enable SOC comparisons to be made between equivalent soil masses, reducing the errors associated with comparisons based solely on horizon depth (Paul et al., 2001b, VandenBygaart and Angers, 2006, Lee et al., 2009). Lee et al. (2009) evaluated the use of three equivalent soil mass (ESM) correction methods and compared them to the use of a fixed depth method for studying C stock changes. In their study, the comparison of mass based SOC changes at a fixed depth (0–15 cm) led to an unrealistic SOC loss of 30% within 6 months following tillage. They suggested that in instances where appropriate soil BD numbers are unavailable the evaluation of SOC concentration changes is most appropriate.

It is ultimately the interaction of management with climatic and edaphic conditions that drive the stabilization or mineralization of SOC (Six et al., 2002, Schmidt et al., 2011). Historical concepts of SOC stabilization have been challenged by advances in modern SOC elucidation techniques (Kleber and Johnson, 2010, Schmidt et al., 2011). In many cases the chemical complexity of SOC has been shown to impart no added protection from microbial degradation and long term persistence has been just as likely for proteins and saccharides as it is for lignin, n-alkanes, long-chain alkanoic acids, and other plant structural tissues (Schmidt et al., 2011). It is clear that SOC stabilization is tightly coupled to temperature, water regime, the quantity and quality of silt and clay minerals upon which C molecules can adsorb, and physical compartmentalization within the soil that can isolate organic compounds from oxygen and microbial decomposers (Six et al., 2002, Kleber and Johnson, 2010, Schmidt et al., 2011). The complexity of SOC stabilization makes questionable a “one size fits all” approach for climate stabilization via SOC sequestration and greenhouse gas mitigation.

We compared the effects of six cropping systems typical of the North Central U.S.A. (3 grain and 3 forage) on the fate of SOC over 20 years using archived and current soil samples to a depth of 90 cm. Our objectives were to assess the effects of row-crop and perennial agriculture, using best management practices, on SOC and to evaluate the relative importance of tillage, perenniality, and crop residue inputs on SOC. We hypothesized that (i) SOC would decrease over time as a result of increased tillage frequency, (ii) perennial forage systems – with limited tillage, deep rooted crops, and manure inputs – would sequester greater amounts of SOC (or lose less) than the annual grain systems, and (iii) within enterprise types (grain and forage), systems with greater species diversity would sequester more C (or lose less).

Section snippets

Site characteristics and experimental design

This study was conducted at the University of Wisconsin's Agricultural Research Station in Arlington, WI (43°18′N, 89°20′W). The soils at the site are classified as Plano silt loam (fine-silty, mixed, superactive, Mesic Typic Argiudolls). These are relatively deep (>1 m), well drained soils with little relief that were formed under tallgrass prairie vegetation in loess deposits over calcareous glacial till. Native SOC (35 g kg−1), silt (720 g kg−1), and clay (215 g kg−1) contents are high as is crop

SOC lost irrespective of cropping system

Type III analysis of fixed effects resulted in a highly significant (p < 0.01) main effect of year (1989 vs. 2009) when considering both SOC concentration and mass data. This significant main effect was due to a loss of SOC (0.6 Mg C ha−1 yr−1 [0–90 cm, Table 3]) at WICST irrespective of cropping system (Table 3). The main effect of cropping system was also significant (p = 0.02) when analyzing SOC concentration data, but was non-significant when SOC mass was analyzed, indicating that the systems did

Discussion

While NT management, application of manure, and inclusion of perennial crops all served to slow the loss of SOC, none of these practices favored C sequestration of over the course of the 20-year trial. The loss of carbon at WICST irrespective of management is consistent with the findings of Bellamy et al. (2005) who showed that SOC was lost across both England and Wales irrespective of land management over a 27-year period. Liu et al. (2011) in a simulation study showed that between 1972 and

Conclusions

Observed changes in SOC at the WICST over the past 20 years highlight the importance of accurate C accounting when drawing conclusions about the sequestration potential of best management practices. While NT management strategies, inclusion of perennial crops, manure, and grass pasture all had beneficial effects on the C stocks at WICST, none of the six systems sequestered atmospheric C when the entire 90 cm profile was considered. These results are consistent with finding at the WICST mirror

Acknowledgments

This work was supported by a USDA-ARS Specific Cooperative Agreement 58-3655-2-0120 with the University of Wisconsin – Madison and the DOE Great Lakes Bioenergy Research Center (DOE Office of Science BER DE-FC02-07ER64494).

References (56)

  • G.R. Sanford et al.

    Effect of dairy slurry application on soil compaction and corn (Zea mays L.) yield in Southern Wisconsin

    Soil Till. Res.

    (2008)
  • J. Simonsen et al.

    Endogeic and anecic earthworm abundance in six Midwestern cropping systems

    App. Soil Ecol.

    (2010)
  • N.H. Batjes

    Total carbon and nitrogen in the soils of the world

    Eur. J. Soil Sci.

    (1996)
  • P.H. Bellamy et al.

    Carbon losses from all soils across England and Wales 1978–2003

    Nature

    (2005)
  • G.A. Buyanovsky et al.

    Carbon cycling in cultivated land and its global significance

    Global Change Biol.

    (1998)
  • R.T. Conant et al.

    Temperature and soil organic matter decomposition rates – synthesis of current knowledge and a way forward

    Global Change Biol.

    (2011)
  • R.T. Conant et al.

    Experimental warming shows that decomposition temperature sensitivity increases with soil organic matter recalcitrance

    Ecology

    (2008)
  • T.H. DeLuca et al.

    Prairie ecosystems and the carbon problem

    Front. Ecol. Environ.

    (2011)
  • J.G. Guzman et al.

    Soil carbon dynamics and carbon budget of newly reconstructed tall-grass prairies in south central Iowa

    J. Environ. Qual.

    (2010)
  • J.F. Hart

    Change in the corn belt

    Geogr. Rev.

    (1986)
  • R.J. Haynes et al.

    Influence of lime, fertilizer and manure applications on soil organic matter content and soil physical conditions: a review

    Nutr. Cycl. Agroecosyst.

    (1998)
  • J.M. Hoenig et al.

    The abuse of power: the pervasive fallacy of power calculations for data analysis

    Am. Statistic.

    (2001)
  • D.R. Huggins et al.

    Corn–soybean sequence and tillage effects on soil carbon dynamics and storage

    Soil Sci. Soc. Am. J.

    (2007)
  • R.B. Jackson et al.

    A global analysis of root distributions for terrestrial biomes

    Oecologia

    (1996)
  • M.K. Jarecki et al.

    Crop management for soil carbon sequestration

    Crit. Rev. Plant Sci.

    (2003)
  • A.E. Johnston

    Soil organic matter, effects on soils and crops

    Soil Use Manage.

    (1986)
  • W. Jokela et al.

    Midwest cropping system effects on soil properties and on a soil quality index

    Agron. J.

    (2011)
  • M. Kleber et al.
    (2010)
  • Cited by (79)

    • Land use-land cover gradient demonstrates the importance of perennial grasslands with intact soils for building soil carbon in the fertile Mollisols of the North Central US

      2022, Geoderma
      Citation Excerpt :

      Maintaining or increasing SOC in light of global change requires a comprehensive understanding of the impacts that current and future agricultural practices have on this valuable resource. Data on changes in total SOC, at depth and over time are difficult to secure, but necessary for accurate assessment of SOC sequestration or loss (Sanderman and Baldock, 2010; Sanford et al., 2012). High variability and large background levels of SOC makes documenting SOC change over time highly uncertain (Necpalova et al., 2014).

    View all citing articles on Scopus
    View full text