Influence of soil and fertilizer nutrients on sustainability of rainfed finger millet (Eleucine coracana) yield and soil fertility in semi-arid Alfisols. Communications in Soil Science and Plant Analysis

Abstract

Productivity of rainfed finger millet in semiarid tropical Alfisols is predominantly constrained by erratic rainfall, limited soil moisture, low soil fertility, and less fertilizer use by the poor farmers. In order to identify the efficient nutrient use treatment for ensuring higher yield, higher sustainability, and improved soil fertility, long term field experiments were conducted during 1984 to 2008 in a permanent site under rainfed semi-arid tropical Alfisol at Bangalore in Southern India. The experiment had two blocks—Farm Yard Manure (FYM) and Maize Residue (MR) with 5 fertilizer treatments, namely: control, FYM at 10 t ha−1, FYM at 10 t ha−1 + 50% NPK [nitrogen (N), phosphorus (P), potassium (K)], FYM at 10 t ha−1 + 100% NPK (50 kg N + 50 kg P + 25 kg K ha−1) and 100% NPK in FYM block; and control, MR at 5 t ha−1, MR at 5 t ha−1 + 50% NPK, MR at 5 t ha−1 + 100% NPK and 100% NPK in MR block. The treatments differed significantly from each other at p < 0.01 level of probability in influencing finger millet grain yield, soil N, P, and K in different years. Application of FYM at 10 t ha−1 + 100% NPK gave a significantly higher yield ranging from 1821 to 4552 kg ha−1 with a mean of 3167 kg ha−1 and variation of 22.7%, while application of maize residue at 5 t ha−1 + 100% NPK gave a yield of 593 to 4591 kg ha−1 with a mean of 2518 kg ha−1 and variation of 39.3% over years. In FYM block, FYM at 10 t ha−1 + 100% NPK gave a significantly higher organic carbon (0.45%), available N (204 kg ha−1), available P (68.6 kg ha−1), and available K (107 kg ha−1) over years. In maize residue block, application of MR at 5 t ha−1 + 100% NPK gave a significantly higher organic carbon (0.39%), available soil N (190 kg ha−1), available soil P (47.5 kg ha−1), and available soil K (86 kg ha−1). The regression model (1) of yield as a function of seasonal rainfall, organic carbon, and soil P and K nutrients gave a predictability in the range of 0.19 under FYM at 10 t ha−1 to 0.51 under 100% NPK in FYM block compared to 0.30 under 100% NPK to 0.67 under MR at 5 t ha−1 application in MR block. The regression model (2) of yield as a function of seasonal rainfall, soil N, P, and K nutrients gave a predictability in the range of 0.11 under FYM at 10 t ha−1 to 0.52 under 100% NPK in FYM block compared to 0.18 under MR at 5 t ha−1 + 50% NPK