An experiment was conducted in a commercial lowland rice-producing area, adopting the minimum tillage system of rice production, followed by two years of fallow, managed, during this time, by cattle production, in order to evaluate the influence on red rice seed bank dynamics. The experimental design was randomized complete blocks, with three replications, in a bifactorial design. Factor A consisted of rice post-harvest management of the area: [T1] - fallow with cattle trampling and [T2] - fallow without cattle trampling. Factor B was the sampling time: [A1] - 1999, [A2] 2000 and [A3] - 2001. The red rice seed bank estimation was made using 12 soil samples by plot, in April 1999, April 2000 and April 2001, using a 10 cm diameter cylinder soil sampler. The sampling depths were 0-1 cm, 1-5 cm, 5-10 cm and 10-15 cm. After being extracted from the soil, the grains of red rice were counted and submitted to a tetrazolium test to estimate seed viability. Cattle trampling did not affect seed distribution in the soil profile and the dynamics of the red rice seed bank. Fallow duration affected the seed bank of red rice; the regression equation fit that best explains the correlation between number of viable seeds and fallow duration in months was the exponential equation: y = 1382.15 exp (-0.1988*x) p<0.05, showing a decrease in the number of seeds within 12 months, from 1,448 to 151 (90% reduction), and 38 (98% reduction) viable seeds per square meter in 24 months of fallow. Regarding the red rice seed burial depths, seed bank reduction was greater on the soil surface in 2000, with no difference between the depths in 2001. The seeds located on the soil surface lost their viability rapidly, with an average of 99% within one year fallow with or without cattle trampling. It can be concluded that cattle trampling does not affect red rice seed bank dynamics in fallow areas. Soil fallow, with or without cattle trampling, reduces in the red rice seed bank.
integration crop/cattle production; weed management; Oryza sativa
