Soil organic carbon stock in pampean soils: changes associated to rotation and tillage

R ALVAREZ

Conferencia; 19th ISTRO Conference; 2012

During the last four decades the cropped area in the Pampas doubled, soybean was introduced in rotations, replacing mainly corn, becoming nowadays the most important grain crop, and no-till adoption by farmers was massively with 80% of the area under this tillage system nowadays. Technological improvements, as the generalized use of fertilizers, lead to yield increases of 100%. Using data from soil surveys performed between 1960 and 1980, a soil organic carbon stock of 4.12 Gt to 1 m depth was estimated. In 2007-2008, 400 sites distributed along the region were sampled for organic carbon evaluation and an artificial neural network model was developed for carbon prediction as a function of vegetation cover and soil use. The model, joined to satellite image classification of the area occupied by common vegetation types and land uses, lead to an estimation of 4.22 Gt of organic carbon. Despite the overall carbon stock did not change during the last decades, decreases were observed in high carbon areas and increases in low carbon ones. Results from 7 long term field experiments were used for fitting a neural network model for estimating organic carbon mineralization. Another neural network model was adjusted to data generated in 116 experiments, which allowed relating crops yield with carbon inputs from residues to the soil. The combination of these two models was used to calculate the soil carbon balance for different rotations and soils. Carbon balances are less negative than 40 years ago or turned out positive in low carbon soils, a consequence of yield increases and higher carbon inputs from straw and roots. The meta-analysis of data from 17 tillage experiments showed that under no-till soil carbon increases by 5 % in rich organic matter soils and 15 % in low organic matter soils at surface. No-till counteracted the impact of soybean on carbon inputs helping carbon level maintenance. The models developed predict that in the following years soil organic carbon of many pampeam subregions will decrease if the proportion of soybean in rotation is not reduced.