A Multivariate Approach to Determine the Effect of Doses and Sources of N, P, and K in Opuntia ficus-indica L. Mill.
Keywords:multivariate analysis, chemical fertilization, prickly pear, young harvested cladodes, nopalitos.
In this work, the effect of doses and sources of nitrogen, phosphorus, and potassium in the production of young cladodes (nopalitos) (cladode sprouts) in prickly pear var. Italiana were studied. The treatments were combinations of doses and sources of commercial fertilizers: T1 = 120 kg N ha-1 (urea), T2 = 120 kg N ha-1 (ammonium nitrate), T3 = 240 kg N ha-1 (urea), T4 = 240 kg N ha-1 (ammonium nitrate), T5 = 100 kg P ha-1 (triple 17), T6 = 100 kg P ha-1 (mono-ammonium phosphate), T7 = 200 kg P ha-1 (triple 17), T8 = 200 kg P ha-1 (mono-ammonium phosphate), T9 = 90 kg K ha-1 (NKS), T10 = 90 kg K ha-1 (potassium sulphate), T11 = 180 kg K ha-1 (NKS), T12 = 180 kg K ha-1 (potassium sulphate), and T13 = control (without fertilizer), which were arranged in a complete randomized block design with four replications. The data were analyzed using univariate and multivariate analysis of variance and multiple comparison means (LSD p=0.05) when the variables showed significant statistical differences among treatments. Results of the univariate and multivariate analysis showed that all variables were significantly different among treatments. The best treatment was T4 [240 kg N ha-1 (ammonium nitrate)], where at least six variables showed the highest values. In general terms, the highest values of the majority of variables were in treatments T1, T2, T3, and T4, which are represented by N; followed by treatments T5, T6, T7, and T8, where the fertilizer was P; treatments T9, T10, T11, and T12, represented by K; and, finally, by the plants of the control (T13). In this sense, results showed a positive effect of N in the production of nopalitos. The three canonical variables explain 94% of the variability due to the treatments, showing that the response of the majority of variables was according the sources of mineral (N, P, and K), where treatments with N were related to the most important canonical variable (CAN1), treatments with P were related to CAN3, and treatments with K were related with CAN2. According to the use of multivariate analysis of variance, we conclude that it was possible to resume and to corroborate our results using this methodology. Although this study is not conclusive, it was a good exploratory method to design the next experiments in the field under the environmental conditions of Baja California Sur using the combination of N, P, and K and different doses of commercial fertilizer.