This study has been developed to evaluate the influence of applying different types and doses of crop residues on potassium (K) mobility in soil columns. Rhodic Haplustox samples were collected at depths of 0.0-0.10, 0.10-0.20 and 0.20-0.30 m and used to create such soil columns, keeping the same profile distribution. The experimental design was randomized with three replications and the tested treatments were organized in a 4x4 factorial arrangement: 4 types of crop residues (brachiaria+sunflower; Mix (cultivated radish+oat+winter vetch); cultivated radish and wheat) and 4 doses of residues (0, 10, 15, and 20 Mg ha -1 ). The grinded and dried residues were applied to the surface of the columns, which were then irrigated with distilled water and incubated for 10 days to stabilize the reactions. After incubation, the soil columns were disassembled and separated into 0. 0. 05-0.10, 0.10-0.15, 0.15-0.20, 0.20-0.25 e 0.25-0.30 m. As amostras de cada camada foram secas, tamisadas e analisadas para determinação do teor de K trocável. Os resultados obtidos indicam que a aplicação superficial de resíduos de culturas altera os teores de potássio trocável do solo, especialmente na camada 0.0-0.05 m. Aumentando as doses de resíduos pode-se aumentar significativa e linearmente os teores de K trocável do solo nas camadas 0.0-0.05, 0.05-0.10, 0.10-0.15 m. Os resíduos da cultura de nabo forrageiro e do Mix (nabo forrageiro + aveia-preta + ervilhaca peluda) foram os que determinaram os maiores aumentos nos teores do K trocável do solo, podendo atingir profundidades de até 0.30 m. O aumento na adsorção do K nas camadas superficiais do solo indica uma menor necessidade de aplicação de adubos potássicos, podendo reduzir os custos de produção e a poluição ambiental. Palavras-chave: Cobertura vegetal, lixiviação, ciclagem de nutrientes, plantio direto, fertilidade do solo

This work aimed to analyze the effect of the variation of the proportions of calcium in relation to the contentes of magnesium and potassium in the cationic capacity of change (CCC) of the soil, seeking to propitiate the  appropriate balance of bases, so that the soybean plants presented good nutritional state and had conditions of resisting to the attack of Phakopsora pachyrhizi Syd. & P. Syd. The experiment was driven in randomized blocks with eight replications and the treatments consisted of doses of Ca:Mg:K in relation to CCC of the soil being:1) without correction of the bases (original soil with 3,8%:6,6%:2,4%); 2) 35:15:5; 3) 45:15:5; 4) 55:15:5; 5) 65:15:5; 6) 75:15:5. The soybean cultivar used in the experiment was BRS 184, sowed in mud vases containing dystrophic Red Latosol (Oxisol) as substrate. The inoculation was realized in the fenologic stadium V4. The disease severity was determined through visual notes considering the percentage foliate area with visible symptoms of the disease, being calculated the Area Under Disease Progress Curve (AUDPC). The data were submitted to the variance analysis (p<0.05) and fitting to regression models. The balance among the nutrientes was analyzed being calculated the index DRIS. The contents of nutrients, the matter dry indexes (MDI) and nutritional balance index (NBI) were inserted in the program ChecarDris for obtaining of the index DRIS. The results appeared that the percentages of 55% of calcium, 15% of magnesium and 5% of potassium, increase larger nutritional balance in the soybean providing to smallest area under disease progress curve (AUDPC), reduction in the severity of the rust (%), besides increase the largest productivity.