Ocorrências de veranicos e seus efeitos sobre o cultivo do consórcio feijão caupi–capim (Panicum maximum) sob dosagens de esterco bovino

Abstract

The semi-arid region covers 86.8% of the State of Ceará, which has the second-largest population in the Northeast of Brazil, with more than 4.5 million inhabitants. A large part of the population in this area is directly linked to agricultural activities and seeks their livelihood on the basis of existing natural resources on or around their own property. For this reason, an understanding of the rainy season and occurring dry spells, and of their influence on the cropping systems used by the rural communities in the semi-arid region of Ceará, such as intercropping Panicum maximum with string beans, is essential for choosing the best planting times and techniques under a rainfed regime. Two experimental studies were therefore carried out. The main objective of the first study was to quantify the occurrence of dry spells of 12, 16 and 20 days in a historical series from 1974 to 2017 in the State of Ceará, in the region of the Central Sertão and in the district of Quixadá, for each month of the rainy season. In addition, an analysis of dry-spell occurrence was carried out in the district of Quixadá, in relation to El Niño and La Niña events. The definition of dry spell was the same as that adopted by Funceme, i.e. a sequence of days when the rainfall depth did not equal or exceed 2 mm day-1. In order to quantify the duration and number of dry spells, rainfall data from the historical series were processed using the PyFuncemeClimateTools framework developed by Funceme. The greatest number of dry spells in the State of Ceará, with a duration of 12, 16 or 20 or more days, occurred in May with 16, 11 and 7 events respectively, representing 57%, 73% and 77% of the events which occurred during the months of the rainy season. During La Niña years, dry spells of 12 or more days were rare in March in Quixadá. Such events did not occur during April. There were no good correlations between the number of monthly events under El Niño or La Niña conditions and SST anomalies, showing that these are not suitable factors for estimating the number of dry spells. In the second study, the principal aim was to verify the effects of dry spells and doses of cattle manure on the productivity of an intercrop of string beans and Panicum maximum grass under semi-arid conditions. The research was carried out on the Bom Jardim farm in the district of Quixadá, Ceará, from May to July 2016. The experiment consisted of four different dry-spell intensities (0, 12, 16 and 20 days, simulated by suppressing the irrigation) and four doses of cattle manure (0, 15, 30 and 45 t ha-1) with four replications, giving a total of 64 experimental units. The experimental design was of subdivided plots. Each experimental unit comprised 8 and 16 planting pits for the grass and beans respectively, with only the central 4 and 8 pits considered as the working area. The variables to be analysed were the number of pods per bean plant, pod length, number of seeds per pod, 100-seed weight, and fresh and dry weight of the bean and grass shoots. The data were submitted to analysis of variance by F-test, and when significant at a level of 1% or 5%, to Tukey’s test (qualitative data, relative to the dry spells) and regression analysis (quantitative data relative to the doses of manure) were carried out. The bean plants developed survival strategies by the end of the dry spells, reducing the number of pods to allow a greater number of seeds per pod. The application of higher doses of manure reduced the effect of the water stress caused by the dry spells in most of the variables under analysis.