The use of cover crops in no-tillage systems (NTS) increases the levels of organic matter and could increase the nitrogen content of the soil, contributing to reduce fertilizers costs. The knowledge of these processes is fundamental for deciding whether cover crops can be effectively incorporated into the agricultural production system. The objective of this study was to evaluate the effect of cover crop species on the levels of nitrate and ammonium in the soil in early upland rice development, as well upland rice yield. A field experiment was performed and treatments consisted of growing rice on five cover crops (Panicum maximum, Brachiaria ruziziensis, Brachiaria brizantha, millet and fallow) in an NTS and two control treatments (Brachiaria brizantha and fallow) under a conventional tillage system, CTS, (one plowing and two disking). The experimental design was a complete randomized block with three replications. The soil samples were collected during a period of six weeks (0, 7, 14, 21, 28 and 35 days in relation to upland rice sowing). The cover crops Brachiaria brizantha, Panicum maximum and Brachiaria ruziziensis in the NTS and B. brizantha fallow incorporated into the CTS favored higher levels of nitrate in the soil. In contrast, B. brizantha and fallow in the CTS and millet and P. maximum in the NTS favored the buildup of high levels of ammonium in the soil. The treatments under the plowed cover crops millet and fallow allowed for a higher upland rice yield. The tillage system and nature of the cover crops could be used to achieve the desired levels and forms of nitrogen in soil. Key words: Brachiaria brizantha, Brachiaria ruziziensis, Panicum maximum, millet, fallow, soil management
ResumoO uso de plantas de cobertura no sistema plantio direto (SPD) aumenta os níveis de matéria orgânica e pode ajudar a aumentar os teores de nitrogênio no solo contribuindo para reduzir os custos de fertilizantes. O conhecimento desse processo é fundamental para que as plantas de cobertura possam ser efetivamente incorporadas aos sistemas de produção agrícola. O objetivo deste estudo foi avaliar o efeito de espécies de plantas de cobertura nos níveis de nitrato e amônio no solo durante o desenvolvimento inicial do arroz de terras altas, bem como a produtividade do arroz. Um experimento de campo foi realizado e os tratamentos consistiram de cultivo de arroz sobre cinco plantas de cobertura (Panicum maximum, Brachiaria ruziziensis, Brachiaria brizantha, milheto e pousio) no SPD e mais dois tratamentos controle (Brachiaria brizantha e pousio) no sistema de plantio convencional, PC (uma aração e duas gradagens). O delineamento experimental foi em blocos completos casualizados com três repetições. As amostras de solo foram coletadas por seis semanas (0, 7, 14, 21, 28 e 35 dias em relação à semeadura do