Multivariate analysis and modeling of soil quality indicators in long-term management systems

Abstract

Soil management systems, as well as the long-term application of nitrogen fertilization, might promote changes in soil quality (SQ). The knowledge of how agronomic practices influence SQ is the main factor in the development of most sustainable management systems. Thus, the aim of this study was to evaluate the influence of long-term management systems on SQ through the analysis of 10 soil quality indicators (SQIs), to select the most sensitive SQIs through principal components analysis (PCA) and to propose a mathematical model that could estimate the activities of enzymes based on SQI values with simple and low-cost procedures in relation to enzyme measurement. Soil samples were collected from three experiments in which soils were used for this purpose over more than two decades. The first experiment consisted of winter fallow and maize seeding as a summer crop in a conventional tillage system (CT) that received nitrogen fertilization at doses of 0, 90 and 180 kg ha−1. The second and third experiments consisted of no-tillage (NT) using maize/maize (NT M/M) and legume/maize (NT L/M) crop rotation, respectively, both using nitrogen fertilization at the same doses as in the first experiment. The no-tillage system with legume/maize crop rotation favored the development of microorganisms and improved the soil quality. The effects of nitrogen fertilization on SQIs varied according to the management system. The microbial respiration (MR), the metabolic quotient (q CO2), total organic carbon (TOC), nitrogen microbial biomass (NMC), urease enzyme activity (UEA), dehydrogenase activity (DA) and amylase activity (AA) were the most efficient SQIs. The adjusted mathematical models presented a good predictive capacity to estimate the urease activity in CT and NT M/M and the amylase activity in the CT system.