Long-term changes in yield and soil fertility in a twenty-year rice-wheat experiment in Nepal

A long-term rice-wheat experiment was conducted at Parwanipur, Nepal, to st udy the effects of organic and mineral sources of nutrients on yield and nu trient status of the soil. Twelve treatments comprising different combinati ons of inorganic N, P, and K; farmyard manure (FYM); and wheat chopped stra w (WCS) were included. On average during a 20-year period, the control plot with an indigenous nutrient supply supported 2.06 Mg rice ha(-1) and 0.69 Mg wheat ha(-1). The application of 100 kg N ha(-1) increased yields, but t he addition of P, K, Zn, and S gave no response, indicating that the soil s upply of these nutrients did not limit yield. The grain yield of rice and w heat was maintained over the years. Soil analyses of the past 6 years (1994 -1999) showed that treatments receiving organic sources of nutrients increa sed total soil C and N from 18% to 62% and 15% to 48%, respectively, compar ed with the NPK treatment. There was a buildup of total P and Olsen P in pl ots receiving FYM. However, total and available soil K were similar in all the treatments. The apparent N and P balances for the rice-wheat system wer e positive in the NPK, FYM, and WCS treatments, whereas the K balance was n egative in all the treatments except with the application of FYM to both ri ce and wheat. Rice and wheat in most years yielded <3.15 Mg ha(-1) and <2.1 6 Mg ha(-1), respectively. These yields are only 40-50% of the potential at tainable yields of this region. Possible reasons for low yields are discuss ed.