Estimating indigenous nutrient supplies for site-specific nutrient management in irrigated rice

Nutrient supplies from indigenous sources (IS) can be estimated by measuring plant nutrient uptake in nutrient omission plots. On-farm experiments were conducted in irrigated rice (Oryza sativa L.) domains of Asia to evaluate relationships of plant N, P, and K uptake with soil tests or grain yield measured in N, P, and K omission (0-N, 0-P, and 0-K, respectively) plots and to develop guidelines for the use of omission plots in site-specific management. Relationships between grain yield or nutrient accumulation and soil tests were scattered. Only 17% of the variation in plant N uptake in 0-N plots was explained by total soil organic C. Extractable Olsen P explained 34% of plant P uptake in 0-P plots, whereas I M ammonium acetate K showed no common relationship with plant K uptake in 0-K plots. With good calibration, indigenous supply of N (INS), P (IPS), and K (IKS) can be estimated from grain yields in omission plots with a precision of about +/-5 to 10 kg N ha(-1), +/-2 to 3 kg P ha(-1), and +/-10 to 20 kg K ha(-1), respectively. Sampling requirements for estimating domain-specific IS values depend on the homogeneity of the domain of interest. For irrigated rice domains of about 100 to 200 km(2), grain yield in omission plots should be measured in at least one high-yielding season in about 10 farms to estimate the domain mean INS, IPS, and IKS. Future research should focus on developing geospatial techniques for delineating fertilizer recommendation domains based on biophysical and socioeconomic characteristics that determine yield potential, IS, and response to fertilizer.