Ammonia Emission in Tropical Cropping Systems.

The main sources of ammonia (NH₃) emissions into the atmosphere are the animal manure and synthetic fertilizers. However, N as NH₃ might be emmited from plant leaves and straw in decomposition, and there is no much information about the magnitude of this process, especially in tropical conditions, wich was our main aim. The experiment has been carried out since 2006 on a clayey Rhodic Hapludox, in Botucatu, state of São Paulo, Brazil. Congo grass (Brachiaria ruziziensis), grain sorghum (Sorghum bicolor) and a mix of both have been planted in the plots during fall/winter season. In the spring, pear millet (Pennisetum glaucum), forage sorghum (Sorghum bicolor) and sunn hemp (Crotalaria juncea) have been planted in sub-plots. Soybean (Glycine max (L.) Merril) has been the summer crop. So we assessed the NH₃–N emission from plants and straw through an entire year (May 2012 to April 2013). To retain the ammonia emitted we used foam absorbers (soaked in solution of glycerin and phosphoric acid) placed 2 cm over the plants/straw. From May to September the systems showed similar variations in rates of NH₃–N emission. In the beginning of October, a first peak of emission was observed, with values reaching to 350 g ha^-1 day^-1 in the brachiaria plots. The second peak fo NH₃–N emission was observed from the beginning of December, being the highest values observed in the Sunn hemp subplots. Both peaks coincided with the desiccation of the crops (fall/winter and spring, respectively). So as is already known, the senescence (in this case induced by the herbicide action) seems to be the phase of the plant cycle with the greatest potential of NH₃ emission. On the other hand, the magnitude of the process seems to be determined by amount of N accumulated in the vegetal tissue once the largest emission rates were observed in the systems in which the plants had the highest content and accumulation of N. The results concerning the total emission of NH₃–N only confirm the variances observed in the daily rates of emission, so the greatest loss of N as NH₃ occurred in the system with Congo grass and Sunn hemp and reached to 19 kg ha^-1 ano^-1. This is a high value, if we consider that there was no N fertilizer supplied. Furthermore, this is an unprecedent result in field conditions and gives us a new view about the N cycling in tropical conditions.