Variability in soil properties influencing pigeonpea (Cajanus cajana L.) yield: a multivariate statistical analysis [version 3; peer review: 2 approved]

Abstract

Aims: The aim of the study was to reveal the variability in soil properties influencing pigeonpea (Cajanus cajana L.) seed yield under semi-arid rainfed condition. Methods: Soils were initially classified into series level and further these series were divided into soil-phase units. For two site years viz., 2018-19 and 2019-20, surface soil samples from each soil-phase unit were collected before sowing of pigeonpea and subsequently crop growth parameters at critical stages were recorded. Results: The principal component analysis with varimax rotation resulted in seven components for both the site years, having eigenvalues greater than one, explained more than 80% of the variability. The step wise linear regression analysis showed that the pigeonpea seed yield was linearly correlated with PC3 (p<0.01), PC4 (p<0.01) and PC7 (p<0.05) of soil properties with R2 = 0.679, during 2018-19. Whereas, during 2019-20, the seed yield was linearly correlated with PC1 (p<0.01), PC3 (p<0.01) and PC6 (p<0.05) with R2 = 0.677. In site year 1, the available P2O5, Fe, Zn, S, Cu, number of pods, surface soil moisture determined the yield. In site year 2, the available K2O, P2O5, Fe, Zn, S, clay, CEC and available water content determined the yield. All these variables together explain variability in yield.

Keywords

• Soil-phase unit
• Soil-plant relationship
• Principal component regression analysis
• eng