CH4 emissions from rice paddies managed according to farmer's practice in Hunan, China

We measured CH4 emissions from rice paddies managed by farmer's practices in Changsha, Hunan Province, China, from 1995 to 1997. During the winter season, rice fields were left fallow under either drained (C-Fallow) or flooded conditions (C-Flood), and planted with either Chinese milk vetch (C-GM) or oil-seed rape (C-Rape). The organic manure produced in the winter (weeds, Chinese milk vetch, or oil-seed rape straw) was incorporated in situ before the early-rice transplanting. Both early-rice and late-rice straws were removed and the soil was not amended with any exogenous organic manure. For 1996 to 1997, the average seasonal CH4 emission for the double rice cropping period was the highest from the plot that was flooded in the winter (103.5 g CH4 m(-2)) and lowest from the plot planted and incorporated with Chinese milk vetch (32.6 g CH4 m(-2)). Precipitation in the winter not only affected growth of green manure, which was incorporated in situ, but might also affect CH4 emissions during the subsequent rice growing period. Therefore, a simple relationship could not be found between the incorporated amount of green manure and CH4 emission. In the plots incorporated with vetch and oil-seed rape straw CH4 emissions were significantly less during the subsequent late-rice period than during the early-rice period. This phenomenon might be attributed to a "priming effect'' of green manure, which exhausted soil labile organic matter. Based on the CH4 flux measurements, the total CH4 emissions from rice fields in Hunan Province during the rice growing season were estimated as 1.56 Tg CH4 in 1996 and 1.06 Tg CH4 in 1997. Large variation of precipitation in the winter would be an important factor controlling the annual variation of CH4 emissions from the treatments.