Concerns about the negative effects of agriculture on soil physical quality and soil organic carbon (SOC) pools have spurred on the adoption of conservation tillage systems in tropical regions. This study aimed to evaluate the long-term impacts (16 th year) of conventional (CT), minimum (MT) and no-tillage (NT) practices and different cover crops (sunn hemp and a bean/ millet sequence) on soil physical properties and SOC content of a corn cultivated Ultisol in the northeast of Brazil. Soil bulk density (Bd), soil penetration resistance (SPR), soil aggregation, and total aggregate-associated carbon (C) (4-2 mm) were evaluated. Tillage practices exerted strong control on soil physical properties and total aggregate-C content but were influenced by cover crop species. Minimum tillage presented the lowest Bd, irrespective of cover crop, while NT with bean/millet sequence resulted in the lowest SPR. However, as regards soil aggregation and total aggregate-C, the results indicated that there were no differences between MT and NT, with both systems presenting greater mean weight diameter (MWD) and total aggregate-C content than CT. Total aggregate-C content in the 0.00-0.05 m soil layer in conservation tillage was increased by the adoption of a bean/millet sequence. Increased mechanical disturbance through CT practices was harmful to Bd, soil aggregation and SOC accrual. Overall, more conservative tillage practices and the proper choice of cover crops might yield greater soil quality. Low intensity of soil disturbances due to the adoption of MT favors soil aggregation and the accrual of SOC in weakly structured soils through increases in contact between organic and mineral particles while not causing significant destruction of soil aggregates.

RESUMOCom o objetivo de avaliar as alterações dos atributos químicos de um Argissolo submetido ao cultivo de milho verde, foram estudados três sistemas de manejo de solo (cultivo convencional, mínimo e plantio direto) combinados com o cultivo de culturas antecessoras (milheto, crotalária, guandu e girassol). O experimento foi instalado nos Tabuleiros Costeiros Sergipanos (10°55'S; 37°11'O) em um Argissolo Vermelho Amarelo, sendo conduzido por 15 anos. Adotou-se o delineamento experimental em faixas (sistema de manejo), onde subparcelas (culturas antecessoras) foram aleatorizadas dentro de cada faixa em três repetições. A produtividade do milho verde e os atributos químicos do solo (0-10 e 11-20 cm) foram determinados. O sistema de cultivo convencional (SCC) apresentou o menor teor de Ca 2+ , Mg 2+ e matéria orgânica do solo (MOS) e o plantio direto (SPD) a maior produtividade da cultura. O cultivo de guandu ou crotalária resultaram no maior teor de MOS. Exceto no SPD, as culturas antecessoras não alteraram as características químicas do solo e a produtividade do milho. Pela análise de componentes principais, o teor de MOS e a produtividade do milho são responsáveis pela maior diferença entre os sistemas de cultivo de milho.Palavras-chave: cultivo mínimo; plantas de cobertura; plantio direto; solos tropicais; Zea mays L. ABSTRACTThree soil management systems (conventional tillage, reduced tillage and no-tillage) combined with predecessor crops cultivation (millet, sunhemp, pigeon pea, sunflower) were studied to evaluate alterations in soil chemical properties of corn cultivated Ultisols. The experiment was set up at the Coastal Plains of Sergipe State (10°55'S; 37°11'W) in a yellow-red Argissol and it has been conducted over the last 15 years. Strip-block experimental design (management) was adopted, where subplots (predecessor crops) were randomized, with three replicates, in each strip-block. Corn yield and soil chemical attributes (0-10 and 11-20 cm) were evaluated. Conventional tillage system (CTS) showed the lowest Ca 2+ , Mg 2+ and soil organic matter (SOM) content and no-tillage (NTS) the highest corn productivity. Pigeon pea and sunhemp cultivation resulted on the highest SOM content. Predecessor crops did not alter soil chemical characteristics and corn yield, except for NTS. Through principal component analysis, SOM content and corn yield are responsible for the greater differences among corn cropping systems.