Site-specific modelling of short-term soil carbon mineralization in central Argentina
•The power model outperformed the exponential model in fitting C mineralization at 28 days.•Random effects on power-model parameters were needed to account for soil variability.•Site-specific mineralization parameters were explained by soil properties.•Short-term C mineralization was higher in soils...
Gespeichert in:
Veröffentlicht in: | Geoderma 2022-01, Vol.406, p.115487, Article 115487 |
---|---|
Hauptverfasser: | , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | •The power model outperformed the exponential model in fitting C mineralization at 28 days.•Random effects on power-model parameters were needed to account for soil variability.•Site-specific mineralization parameters were explained by soil properties.•Short-term C mineralization was higher in soils with higher organic matter and fertility.
Soil organic carbon (SOC) mineralization is essential for nutrient cycling; however, it also increases carbon emission as CO2 to the atmosphere (CO2-C). Therefore, detailed information is required to design land use practices promoting positive C balances and soil health. Short-term SOC mineralization is sensitive to soil properties. Short-term soil C mineralization at a landscape scale has been poorly describe, despite the need to understand the variability of C dynamics among environments. The objectives of this work were to 1) estimate the parameters of short-term soil C mineralization kinetics in different locations within a region and 2) explore correlations among site-specific kinetic parameters and edaphoclimatic properties of sampled sites. The study involved 150 sites with different soil and climatic characteristics sampled from a 165,000 km2 area in central Argentina. CO2-C data were obtained from 28-day laboratory incubation experiments of the 150 soil samples. We tested three models (exponential, hyperbolic and power), which depend non-linearly on their parameters, to describe the population average cumulative CO2-C. Using nonlinear mixed models, we tested the significance of random site effects to explain variability in curve parameters among soils. Random forest was used to explain C mineralization parameters from edaphoclimatic properties. The best fit was provided by the power model with random effects on two parameters: initial C mineralization rate and mineralization rate coefficient. The results indicate heterogeneous C mineralization processes across central Argentina. The observed cumulative CO2-C after 28 days of soil incubation ranged between 250 and 1693 mg C kg−1 soil with an average of 826 mg C kg soil−1. Lower amounts of CO2-C (800 mg C kg−1 soil) are expected at 28 days of mineralization in sandy loam soils with lower organic matter than in rich loam soils (more than 1000 mg C kg−1 soil). A relational study explaining site-specific C mineralization from soil, climatic and management variables indicated a significantly higher short-term C mineralization in soils with high organic matter and so |
---|---|
ISSN: | 0016-7061 1872-6259 |
DOI: | 10.1016/j.geoderma.2021.115487 |