Investigating the effects of Nitrogen and Phosphorus Fertilizer Application on Yield Attributes, Grain Yield and Quality of Rain Fed Rice (NERICA-3) in Gambella, Southwestern Ethiopia

A field experiment was carried out during the 2008 and 2009 crop seasons at Imla, Gambella Zuria District, Gambella, southwestern Ethiopia, to establish the application rates of N and P fertilizers for rice variety NERICA-3 (Oryza sativa x Oryza glaberrima). Rice can produce grain yield as high as 10-18 t ha-1 in countries advanced in its cultivation . However, its productivity in Ethiopia in particular (2.31 t ha-1 of paddy rice) and in Africa at large is much below its world average (4.35 t ha-1 of paddy rice) due to improper crop management practices. The treatments consisted of factorial combinations of four rates each of N (0, 46, 92 and 138 kg N ha-1) and P (0, 23, 46 and 69 kg P ha-1) laid down in a randomized complete block design (RCBD) with three replications. The effects of year showed significant (P \(\le\) 0.05 for some and P \(\le\) 0.01 for most) differences for leaf area index (LAI), 1000-grain weight, plant height, panicle length and grain qualities. Similarly, the effects of N were significant (P \(\le\) 0.01) for productive tillers plant-1, grains panicle-1, LAI, plant height, panicle length, grain yield, crude protein, ether extract and crude fiber. Growth, yield, and quality components did not differ significantly (P > 0.05) due to application of P except crude fiber. Conversely, the effects of year by N interaction were significant (P \(\le\) 0.05 and/or P \(\le\) 0.01) for LAI, panicle length, grains panicle-1, crude fiber, plant height and ether extract. Effects of year by P interaction were significant only for crude fiber whereas the interaction effects of N by P and year by N by P on growth, yield and quality parameters were not statistically significant (P > 0.05). Rice grain yield increased from 3.54 to 5.90 tons per hectare (t ha-1) with increase in level of N from the control to 92 kg N ha-1 but decreased with further increase of N. In conclusion, sensitivity analysis on coexisting changes in field prices of inputs and rice grain (±15%) showed that 92 kg N gave the highest (681.53%) marginal rate of return (MRR) followed by 23 kg P ha-1 (117.44%). Therefore, application of 92 kg N ha-1 to improve grain yield of rain fed NERICA-3 rice might be more profitable even under risky market situations in and around the study area.