Influence of continuous and intermittent water regimes on methane emissions from irrigated rice crops in southeast Brazil.

Irrigated rice crops in Brazil correspond to 35% of the total rice area and the preferred irrigation method is continuous flooding. In this study two water management systems ? continuous and intermittent flooding ? were monitored for the purpose of comparing their methane emission potential. Intermittent flooding was characterized by alternating flooding and draining periods during the rice cropping cycle. In both systems nitrogen fertilizers were applied twice (NPK and urea alone). Methane fluxes were determined using a closed chamber method (boxes 60x60 cm), and analyzed by GC-FID. Samples were collected from 4 boxes (2 boxes for each water management system) twice a week, for a weekly total of 46 samples. Measurements of air and soil temperature, soil and water pH, redox potential, plant biomass were done. After 104 days the accumulated fluxes resulted in a mean methane emission of 21.1 9 g CH4 m-2 d-1 under the continuous flooding regime and 23.9 g CH4 m-2 d-1 under the intermittent flooding regime. After the two N fertilizer application, methane fluxes decreased in all boxes, with a more marked reduction occurring in the intermittent regime stands. Total accumulated methane emissions, as well as grain yield, were evaluated for each water management system at the end of the cropping season, showing 13% higher productivity in the field under intermittent flooding system. The objectives of the present study were (i) to improve methane emission estimates and compare the emission potential of traditionally managed crop systems and (ii) to study the possibility of minimizing water consumption and the influence on the methane fluxes without deleterious effects to the crop