Land Use Effects on Soil Organic Carbon Fractions in the Pampas of Argentina Determined via Inverse Analysis of Carbon Mineralization Data

ABSTRACT

Understanding the long-term effects of land use on soil organic carbon (SOC) is of interest due to the possibility of using soils as carbon (C) sinks. We determined how the C levels in the labile (C₀) and stable (C_r) fractions of the organic matter of soils of the Pampas were affected by land use, applying the inverse analysis methodology. Soil samples (0–50 cm) taken from 198 sites under different land uses (forest, grassland, and cultivated soil) were incubated for six months, adjusting a one pool exponential model to C mineralization data. The model allowed a good estimation of C₀ (R² = 0.96–0.98) and C_r was calculated by difference with SOC. The stocks of C₀ and C_r were calculated on an equivalent mass basis (5000 t soil ha⁻¹) for comparisons. About 80% of the SOC is found in the C_r pool in the Pampas soils. Both C₀ and C_r were higher in forest and grassland than in cultivated soils. Soil cultivation produced an average reduction of 30% of the C₀ pool and 18% of the C_r pool in relation to grassland soils. Consequently, most of the SOC lost due to agriculture came from the C_r pool. In mixed production systems, no differences of the C pools were detected between the crop or pasture phase of the rotation. The parameters C₀ and k (mineralization rate constant) of the exponential model were negatively correlated, so the comparison of C₀ between different land uses must be careful. The k parameter was similar among land uses when the comparison was made for similar values of C_0, suggesting that land use affected the size of the C pools not their stability. Forest and grasslands were effective systems for sequestering C in stable forms. Expanding the cultivated area in the Pampas should be avoided if it is desired not to emit C into the atmosphere from the soil.

KEYWORDS:

• Land use
• soil organic carbon
• carbon mineralization
• inverse analysis

Acknowledgments

This research was granted by the Universidad de Buenos Aires (Project 20020130100484BA) and CONICET (Project 084, period 2014–2016).

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by the Consejo Nacional para Investigaciones Científicas y Tecnológicas [Project 084, period 2014–2016]; University of Buenos Aires [Project 20020130100484BA].