Under rainfed upland situations, rice grows in an environment prone to N losses due to volatilization and runoff. Farmers in eastern India hesitate to invest more in N inputs because of stress (biotic and abiotic)-related risks. Earlier studies (Sinha et al 1994, Singh et al 2002) recommended N application in two splits (20 and 40 d after rice emergence) in upland rice. We compared the response of upland rice variety Vandana to two split N applications: two N splits with basal and two N splits only during four consecutive wet seasons (2000-03).
The on-farm experiment used a randomized complete block design involving 11 farmers' fields under rainfed conditions in villages Khorahar, Handio, and Sakhia in Hazaribag, Jharkhand, India. The 11 fields were treated as replications. Soil in farmers' fields varied from silt loam to sandy loam, with pH 5.9-6.1, 0.22-0.39% organic C, 2.67-5.31 ppm double acid-soluble P, and 131-176 ppm available K. Fields were plowed once in the off-season (December-April) and three times before seeding. The recommended levels of P (13 kg ha-1) and K (16 kg ha-1) were applied in the form of single superphosphate and muriate of potash, respectively. Short-duration (95 d) variety Vandana was used as the test variety.