Conservation Agriculture in cereal systems of South Asia: effect on crop productivity and carbon-based sustainability index Jat ML 1 , Jat RK, Gupta R, Gopal R International Maize and Wheat Improvement Center (CIMMYT), NASC Complex, New Delhi-110012, India; M.Jat@cgiar.org Key words: conservation agriculture, cereal systems, carbon sustainability index Introduction Maize, wheat and rice, the major cereals grown as monoculture or in sequence, contribute the bulk of food and incomes in rural regions of South Asia. The large increases in food production are one of the greatest achievements of the second half of the 20 th century, but sustaining these gains is a major challenge. Inappropriate management practices and intensive cropping in the past have led to the twin challenges of resource depletion and decelerating productivity growth of cereal crops. The efficiency and sustainability of a production system depends on system-based management optimization of crop yields, economic benefits, and environmental impacts. The long term sustainability of a cropping system depends on its carbon inputs, outputs and carbon use efficiency which indicate long- term sustainability in terms of yield, environment and ecology. Similarly, efficient utilization of carbon-based resources mitigates the increasing level of CO 2 in the environment. Most of the indices used by researchers in the past to evaluate the sustainability of any cropping systems have been based on economic yield or yield sustainability, without giving due emphasis to environmental issues. In this study, three major cereal systems (rice-wheat, rice- maize and maize-wheat) were evaluated under conservation agriculture practices and compared with conventional tillage in terms of yield, carbon (C) inputs and outputs, and a carbon based sustainability index. Materials and Methods Three tillage treatments (no-till on the flat, permanent raised beds, conventional till) and 3 cropping systems (rice-wheat, rice- maize, and maize-wheat) were implemented in large plots (1000 m 2 ) at Rajendra Agricultural University, Samastipur, and also in on-farm (farmer participatory) trials, in Bihar, India. The different crop sequences were initiated in different years - rice-wheat (2007), rice-maize (2008), maize-wheat (2009) – resulting in datasets for 4, 3 and 2 years, to date, respectively. All crops except transplanted rice were planted using a multi-crop, multi-purpose no-till seed-cum-fertilizer planter. The permanent beds were of 67cm width (furrow-to-furrow) with 2 rows of rice or wheat and one row of maize. In the conventional till system, the maize and wheat were planted after 4-dry ploughings, while the rice fields received dry tillage followed by wet tillage (puddling) and transplanting of 23 day old seedlings. The residue management was common irrespective of tillage wherein partial (30-cm anchored) residues of monsoon crops (rice and maize) were recycled and in winter crop (wheat) only 10-cm stubbles were recycled. Carbon sustainability index (CSI), the ratio of the difference between C output and input to the C input (Lal, 2004), was used for analysis of the sustainability of the cropping systems x tillage treatments. C input was calculated as the sum of the C equivalent of all inputs i.e. tillage operation, fertilizer, seed, pesticide, harvesting, threshing etc. C output was computed as the sum of the C equivalent of grain, straw and root biomass produced by the crop. C