||In the intensive rice-wheat system (RWS) of the western India, direct dry seeding of rice, zero tillage in both rice and wheat, and residue recycling from rice can greatly reduce water and labour input, eliminate residue burning, and potentially conserve natural resources. The objectives of this 2-year field study conducted at the Project Directorate for Cropping Systems research Modipuram, India in 2005 to 2007 were to determine the effects of tillage, crop establishment methods and residue on crop growth, physiology, grain yield, water productivity, and economic profitability in RWS. The treatments included six combinations of rice establishment and tillage in rice with and without Sesbania co-culture. In the following season wheat was raised as zero-till (ZTW) in all the treatments and the residue from previous rice was either removed or retained on the soil surface without incorporation. Compared to conventional till puddled transplanted rice (CT-PTR), rice yields in dry direct-seeded rice (DSR) averaged 9.9 % lower with zero till (ZT) and 7.3% lower with CT. Rice yields were similar from CT-DSR and ZT-DSR. The Sesbania co-culture in rice increased mean grain yield by 5% compared to no Sesbania. DSR under CT and ZT was characterized by more panicle number per m2 but less grain number per panicle, and lower harvest index compared with CT-PTR. Mean wheat grain yield was significantly higher from crop planted after ZT-DSR (5.03Mg ha-1) compared to sown after CT-PTR (4.74t ha−1), and CT-DSR (4.86tha-1). Higher leaf water potential and increased photosynthesis rates were recorded in wheat under residue mulch compared to no mulch. Canopy temperature in wheat during the grain filling stage was 0.6-1.5oC lower under residue retention compared to residue removal. The normalized vegetative index (NDVI) adequately described the effect of residue mulch on the growth of both rice and wheat crops. Residue retention increased mean grain yield of wheat by 9.9% compared to removal of residue. On an average, DSR used 7-13.9% less water compared to CT-PTR. Water productivity of wheat was highest (0.91kg grain m–3 water) in ZT-DSR/ZTW+R and lowest (~0.77kg grain m–3 water) in CT-PTR/ZTW and CT-DSR/ZTW. Overall, DSR/ZTW+R had higher water productivity and net returns than CT-PTR/ZTW in the RWS. Our study showed that conservation agriculture-based crop management practices (ZT-DSR/ZTW+R) could provide a better alternative to the conventional RWS to sustain high productivity while significant savings in water, labour and production costs under changing climate.