Experiments were conducted in two sugarcane fields, one with low weed pressure (plot 1) and the other with high weed pressure (plot 2). Four treatments were established in each study: green harvest that left plant mulch on soil surface (VCR), green harvest without plant mulch on soil surface (VSR), burning the crop before harvest without soil disturbance (QSL) and burning the crop before harvest with soil disturbance (QCL). Evaluations were made 75 days after harvest. The evaluations were conducted separately in rows and between rows. In addition, sugarcane population and height were measured. Differences in weed populations were not observed for the field with low weed pressure (plot 1). In the field with high weed pressure (plot 2), weeds were favored by burning and soil disturbance caused by fertilizer incorporation. Thus the treatments QSL and QCL showed the highest percentage of surface coverage. Also, the QCL treatment had the greatest weed populations resulting in reduced sugarcane population and height. Purple nutsedge (Cyperus rotundus L.) and itchgrass (Rottboellia cochinchinensis (Lour.) Clayton) had the greatest percentage surface coverage, especially in QCL. Weed populations were larger in rows than between rows. Apparently in rows, weeds found better conditions for germination and nutrient uptake.