Spatial and temporal variability of Mehlich-1 extractable Fe, Mn and Zn over a rice field as a function of lime amendment

Abstract

This study compares the effect of lime additions on the spatial variability of Fe, Mn and Zn extracted by Mehlich-1 during three different growth stages from an acid paddy soil, a Typic Plintacualf, in Corrientes, Argentina. Field trials were set up involving three treatments: control, without lime addition, plus two different dolomite doses of 625 and 1250 kg ha⁻¹. Soil was sampled first before sowing in aerobic conditions and then two more times in anaerobic conditions, i.e. by bunch formation and flowering. Ninety-six samples per plot were taken per lime treatment and sampling period, using a nested sampling strategy. Liming significantly increased extractable Fe and Mn, but decreased extractable Zn. The spatial variability of the studied soil properties was assessed using semivariogram analysis and examination of kriging maps. Models were fitted to experimental semivariograms for 27 data sets, i.e. three different soil properties, each of them sampled in three treatments and during three dates. Soil extractable and Fe, Mn and Zn exhibited a rather strong spatial dependence, as nugget variance was either null or a small proportion of the total variance, and this all over the three different study periods and for the three lime treatments. Geostatistical analysis provided insight into possible processes responsible of the observed spatial variability patterns within the rice soil. Kriging was useful in mapping soil micronutrient variability allowing identifying microrregions with high or low Fe, Mn and Zn concentrations, which showed the presence of small scale variability. These findings indicate the potential for applying the principles of precision agriculture to control spatiotemporal variability in rice fields.