营养物
肥料
护根物
农学
土壤肥力
土壤水分
干物质
数学
化学
环境科学
生物
土壤科学
有机化学
作者
Aliou Saïdou,B.H. Janssen,E.J.M. Temminghoff
标识
DOI:10.1016/s0167-8809(03)00184-1
摘要
Four on-farm experiments examined whether modest applications of fertilizers in combination with prunings from native agroforestry trees would be an alternative to maintain the fertility of ferralitic soils in Benin. An application of about 1.9 t ha−1 dry matter of mulch of Senna siamea combined with 30 kg N ha−1, 22 kg P ha−1 and 25 kg K ha−1 as compound fertilizer was compared with (1) 60 kg N ha−1, 43 kg P ha−1 and 50 kg K ha−1 as compound fertilizer alone, (2) mulch of S. siamea alone (about 3.2 t ha−1 dry matter), and (3) a control treatment. Criteria were soil properties, yields, nutrient uptakes, and nutrient budgets. Application of sole mulch had no significant effects (P>0.05) on maize yields, while combined application of prunings and NPK fertilizers or sole NPK increased yields significantly (P<0.05). The most limiting nutrient was P. The local maize cultivar was efficient in P uptake, but not in internal nutrient utilization efficiency; mulch increased significantly the internal P utilization efficiency (P<0.05). Soil properties were interpreted with the QUEFTS (quantitative evaluation of the fertility of tropical soils) computer program. The predicted and measured yields were almost the same for maize without NPK. The measured responses to NPK were much lower than the responses calculated by QUEFTS. The calculated nutrient budgets were split into balances for available nutrients and for those not immediately available (NIA). Nutrient budgets were negative for the control and sole mulch treatments, and positive for the NPK treatments. Mulch improved the balances of NIA nutrients. The present experiment could not prove that combining NPK with mulch is the best option for sustainable agriculture. It may be more economical to apply lower rates of fertilizer to local maize than those applied in the two NPK treatments in the present study.
科研通智能强力驱动
Strongly Powered by AbleSci AI