Empirical model of the relation of output value to nitrogen, phosphorus, potassium and water during tobacco cultivation

Chinese. 为优质烟叶生产提供施肥量依据，防止化肥的过量施用从而造成的环境污染及烟叶品质下降。以经典统计学和地统计学为分析工具，在对试验地土壤养分空间变异性分析的基础上进行二次回归旋转组合设计试验。建立了氮、磷、钾肥及水分4因子与烟草产值之间的经验模型，并对各因子之间的交互作用进行了分析，对模型进行了优化。单因子效应分析表明，随着施氮量和土壤相对含水量的提高，产值呈先升后降的趋势，随着施磷量与施钾量的提高，产值呈上升趋势。双因素交互效应分析表明，氮肥和水分、磷肥和水分、钾肥水分、氮、磷肥、氮、钾肥间互作中都存在一个值域，低于这个值域时氮肥和水分、磷肥和水分、钾肥和水分、氮、磷肥、氮、钾肥间都表现为协同促进作用，高于这个值域时则都表现为拮抗作用。而钾肥和磷肥间一直表现为协同促进作用。模型优化结果表明，大田试验适宜的施氮量为36.72～45.65 kg N•ha-1，施磷量为56.53～71.52 kg P2O5•ha-1，施钾量为240.78～250.94 kg K2O•ha-1，土壤相对含水量为75.2%～80.8%。

English. The present experiment was conducted to study the reasonable fertilization for high quality tobacco leaves and to prevent from the environmental pollution and the quality decline of tobacco leaves resulted from over fertilization. Quadratic general rotary unitization design was used to study spatial heterogeneity of soil nutrients with the classical method and geostatistics. An empirical model of the relations of tobacco output to nitrogen, phosphorus, potassium and soil water content was established and optimized to analyze the interaction among the tested factors. Single-factor analysis showed that the output of tobacco leaves increased and then decreased steadily with the increase of the nitrogen application and the soil water content, but it always kept going up with the increase of the application of phosphorus and potassium. The first order interaction analysis indicated that there existed an interaction range between nitrogen and water, phosphorus and water, potassium and water, nitrogen and phosphorus, nitrogen and potassium. The synergistic and antagonistic effects between fertilization and watering were shown in the lower and higher range, respectively. But the synergistic effects always existed between phosphorus and potassium. The optimal amount in the field from the optimized model was 36.72 kg•ha-1 to 45.65 kg•ha-1 for N, 56.53 kg•ha-1 to 71.52 kg•ha-1 for P2O5, 240.78 kg•ha-1 to 250.94 kg•ha-1 for K2O, 75.2% to 80.8% for the soil water content, respectively.