Influence of Previous Crop on Durum Wheat Yield and Yield Stability in a Long-term Experiment

Long-term experiments are leading indicators of sustainability and serve as an early warning system to detect problems that may compromise future productivity. So the stability of yield is an important parameter to be considered when judging the value of a cropping system relative to others. In a long-term rotation experiment set up in 1972 the influence of different crop sequences on the yields and on yield stability of durum wheat (Triticum durum Desf.) was studied. The complete field experiment is a split-split plot in a randomized complete block design with two replications; the whole experiment considers three crop sequences: 1) three-year crop rotation: sugar-beet, wheat + catch crop, wheat; 2) one-year crop rotation: wheat + catch crop; 3) wheat continuous crop; the split treatments are two different crop residue managements; the split-split plot treatments are 18 different fertilization formulas. Each phase of every crop rotation occurred every year. In this paper only one crop residue management and only one fertilization treatment have been analized. Wheat crops in different rotations are coded as follows: F1: wheat after sugar-beet in three-year crop rotation; F2: wheat after wheat in three-year crop rotation; Fc+i: wheat in wheat + catch crop rotation; Fc: continuous wheat. The following two variables were analysed: grain yield and hectolitre weight. Repeated measures analyses of variance and stability analyses have been perfomed for the two variables. The stability analysis was conducted using: three variance methods, namely the coefficient of variability of Francis and Kannenberg, the ecovalence index of Wricke and the stability variance index of Shukla; the regression method of Eberhart and Russell; a method, proposed by Piepho, that computes the probability of one system outperforming another system. It has turned out that each of the stability methods used has enriched of information the simple variance analysis. The Piepho’s probability method, moreover, abridges in effective way the analysis of variance results, supplying precise indications about the influence of crop sequence on quali-quantitative productive variables; in particular, wheats in three-year crop rotation (F1 and F2) have higher probability to obtain higher qualitative and quantitative productions than one in one-year crop rotations (Fc+i and Fc), so as wheat in one-year crop rotation with catch crop vs. wheat monoculture.