INVESTIGADORES
ALVAREZ Roberto
artículos
Título:
Land use effects on soil carbon in the Argentine Pampas
Autor/es:
B GONZALO; R ALVAREZ; J DE PAEPE; C CARIDE; R CANTET
Revista:
GEODERMA
Editorial:
ELSEVIER SCIENCE BV
Referencias:
Lugar: Amsterdam; Año: 2013 vol. 192 p. 97 - 110
ISSN:
0016-7061
Resumen:
Our objective was to establish the pattern of variation of soil organic (SOC) and inorganic (SIC) carbon stored in surface and deep soil layers of the Argentine Pampas as affected by environmental conditions and land use. Eighty two farms, widespread over the region, were used for the study. At each farm paired treatments were sampled representing common land uses: trees, uncropped controls, seeded pastures, cropped fields and periodically flooded areas. Bulk density, SOC, SIC, texture, pH and electrical conductivity were determined to 1 m depth. Rainfall and temperature were obtained from climatic records. Significant differences were detected between treatments in SOC contents. Average SOC stocks to 1 m were: 131 t ha-1 under trees > 101 t ha-1 in uncropped control > 90 t ha-1 in pastures = 86 t ha-1 in cropped field > and 70 t ha-1 in flooded sites. Compared with uncropped controls, SOC was significantly different in all soil layers under trees, to 75 cm depth in flooded sites and to 50 cm in pastures and cropped soils. Agriculture determined a reduction of 16 % of SOC to 50 cm in sampled sites. In the 50-100 cm depth a decrease of 9 % was observed, though not significant. The stratification pattern of SOC in depth was not affected by the treatments; implying that land use impacted the SOC sequestered in soil, but not its allocation in depth. SIC accounted for one third of total soil carbon, average SIC stock was 50 t C ha-1 to 1 m. Both, its stock and distribution in the profile were not affected by the treatments; with greater SIC stocks founded in deep soil layers.  An artificial neural network model was developed that allowed the estimation of SOC (R2= 0.64) based on climate, soil properties and land use. The model, linked to information from satellite image classification, was used for the estimation of present SOC stock of pampean soils, which accounted for 4.22 ± 0.14 Gt in an area of 48.2 Mha. Using soil surveys performed during the 1960-1980 period we estimated a SOC stock of 3.96 ± 0.22 Gt. Consequently, no change of total SOC stock seems to be produced in the last decades in the region. At smaller scale, counties with SOC content greater than 95 t ha-1 to 1 m depth lost carbon; increases prevailed below this threshold. Apparently, SIC reservoirs seem have not change during the last decades.