基于GIS和丰缺指标法的区域施肥管理体系的构建

  【目的】  采用GIS定位和养分丰缺指标相结合的方法，研究西辽河平原耕地土壤碱解氮、有效磷、速效钾的空间分布特点，建立直观、精准的推荐施肥系统。  【方法】  西辽河平原包括科尔沁区、开鲁县、奈曼旗、科尔沁左翼后旗、科尔沁左翼中旗。以该区域的15421个土壤测试数据和143个玉米“3414试验”数据为基础，运用GIS方法计算耕地土壤碱解氮、有效磷、速效钾的空间插值结果，结合土壤丰缺指标，将西辽河平原耕地土壤碱解氮、有效磷、速效钾养分分区，确定不同土壤氮磷钾养分组合下的施肥量，并制作玉米氮磷钾分区施肥图。  【结果】  西辽河平原耕地土壤养分空间分布不均，大部分土壤碱解氮含量较低，有效磷、速效钾含量中等。耕地土壤有效氮、磷、钾养分含量呈低氮–中磷–中钾 (碱解氮 ≤ 69.5 mg/kg、有效磷4.18～10.40 mg/kg、速效钾90.3～152.5 mg/kg)、中氮–中磷–中钾 (碱解氮69.5～107.9 mg/kg、有效磷4.18～10.40 mg/kg、速效钾90.3～152.5 mg/kg)、中氮–中磷–高钾 (碱解氮69.5～107.9 mg/kg、有效磷4.18～10.40 mg/kg、速效钾 > 152.5 mg/kg) 的面积占比分别为42.4%、14.1%、13.5%。土壤碱解氮、有效磷、速效钾含量 (x) 与经济最佳施肥量 (y) 之间存在较好的对数关系，氮施肥模型y = –102.5ln(x) + 617.22，磷施肥模型y = –36.11ln(x) + 174.1，钾施肥模型为y = –25.89ln(x) + 180.96。不同土壤养分组合种植玉米所需的经济最佳N–P₂O₅–K₂O用量 (kg/hm2) 组合分别为202–110–58、168–107–57、164–102–47。  【结论】  土壤碱解氮、有效磷、速效钾含量与相对产量存在较好的对数函数关系。西辽河灌区土壤有效氮磷钾丰缺指标的划分以75%和85%作为相对产量的分级区间较为合适。在土壤有效氮磷钾养分大数据的支撑下，采用GIS和丰缺指标法相结合的方法，可以准确快捷地构建区域施肥管理体系。

Construction of regional fertilization system based on GIS and nutrient abundance index

  【Objectives】  We assessed available soil nutrients using a nutrient abundance index and graded the spatial distribution of the former based on GIS and soil nutrient grading in Xiliao River Plain. Further, we proposed fertilizer recommendations across the spatial layout.  【Methods】  Xiliao River Plain in Inner Mongolia covers Horqin District, Kailu County, Naiman Banner, Horqin Left Back Banner, and Horqin Left Middle Banner. Based on 15421 soil test data and 143 “3414” maize trail data, we calculate the spatial interpolation results of alkaline N, available P, and readily available K in the cultivated soil using GIS. Combining GIS with the method of soil abundance index, the nutrient management zones of soil alkaline N (AN), available P (AP), and readily available K (AK) were divided in Xiliao River Plain. The fertilizer uptake of different soil NPK nutrient combinations was determined. The map of N, P, and K zonal fertilization of maize was developed.  【Results】  The spatial distribution of soil nutrients in the cultivated soil was uneven. The content of soil AN was low, and the contents of soil AP and AK were medium in Xiliao River Plain. The most important combinations of soil AN–AP–AK were low (N)–medium (P)–medium (K), medium (N)–medium (P)–medium (K), and medium (N)–medium (P)–high (K). The area ratio of each combination was 42.4%, 14.1%, and 13.5%. There were better logarithmic relationships between the soil AN, AP, and AK content (x) and the economic optimal fertilizer rate (y). Nitrogen optimal fertilization model was y = –102.5ln (x) + 617.22, phosphorus optimal fertilization model was y = –36.11ln (x) + 174.1, and potassium optimal fertilization model was y = –25.89ln (x) + 180.96. Under the different soil nutrient combinations, the optimal dosage combinations of N, P₂O₅, and K₂O for maize were 202–110–58 (kg/hm2), 168–107–57 (kg/hm2), and 164–102–47 (kg/hm2).  【Conclusions】  The relative yield of maize has a good logarithmic relationship with soil AN, AP, and AK content. The simulation effects of the function between the relative yield of maize and soil nutrients differ under different methods.When dividing the abundance index of soil alkaline nitrogen, available phosphorus, and readily available potassium in the Xiliao River Plain, grading intervals of the relative yield of 75% and 85% are suitable. Under the big data support of soil nutrients, the regional fertilization management system can be established accurately and rapidly by combining the GIS and abundance index methods.