Sensitivity of different soil quality indicators to assess sustainable land management: Influence of site features and seasonality

Abstract

The turnover rate of labile organic fractions varies continuously due to different soil uses and managements, weather conditions and sampling time. The aim of this study was to quantify the effect of different agricultural management, season and soil type on soil organic carbon (SOC) and its different fractions. The study was conducted on four sites located in the Argentinean Pampas. In each site, three treatments were defined: Good Agricultural Practices (GAP), Poor Agricultural Practices (PAP) and Natural Environment (NE). During two consecutive years (2010 and 2011) and at two different times (February and September) undisturbed soil samples were taken at 0-20 cm depth. Variables assessed included: SOC and its organic fractions: coarse (POCc) and fine (POCf) particulate organic carbon, SOC associated with a mineral fraction (MOC), total (CHt) and soluble (CHs) carbohydrates, bulk density (BD), and large pores (P>30). Also, indices associated with soil and management variables were determined. SOC reductions caused by agricultural practices were mainly from POCc. This fraction represented 34-52% and 50-74% for PAP and GAP, respectively, of the observed in NE. The carbon pool index (CPI) shows that agricultural treatments induced greater variations in all the labile organic fractions compared with SOC and MOC. In turn, the magnitude of variability was different among fractions, where temporal fluctuations increased according to the following order MOC < SOC < POCf ≤ CHt < CHs ≤ POCc. Independently of the soil type, the CPI was a sensitive indicator of soil quality in these systems under no-tillage. The multivariate analysis has proven to be an efficient analytical methodology for the identification of soil indicators that respond to agricultural practices, in which chemical properties (POCf and CHt), physical (BD and P>30), and indices (SOC: clay, structural index and intensification sequence index) were the variables that best explained the total variance of information of the four sites. Therefore, these indicators/indices should be included in any minimum data set for evaluating the agricultural soil quality under no-tillage in the studied area.