欢迎访问《中国农学通报》，
中国农学通报

中国农学通报 ›› 2010, Vol. 26 ›› Issue (8): 179-188.

• 资源 环境 生态 土壤 气象 • 上一篇    下一篇

PI模型在东北松嫩黑土区土壤生产力评价中的应用

段兴武 谢云 张玉平 刘冰   

  • 收稿日期:2009-10-09 修回日期:2010-01-11 出版日期:2010-04-20 发布日期:2010-04-20
  • 基金资助:

    维持可持续土地生产力的定量标准：容许土壤流失量

Applied PI model in soil productivity assessment of Song Nen black soil region in Northeast China

Duan Xingwu, Xie Yun, Zhang Yuping, Liu Bing   

  • Received:2009-10-09 Revised:2010-01-11 Online:2010-04-20 Published:2010-04-20

PDF (PC)

1038

摘要：

摘 要：【研究目的】综述土壤生产力指数模型（Soil Productivity Index，简称PI）的原理、方法以及在国外的运用，探讨其在松嫩黑土区的适宜性以及改进的可能。【方法】在系统介绍PI模型的基础上，以东北松嫩黑土区为研究区，利用研究区17个土壤剖面理化性质数据和作物产量调查数据，比较分析了表土评价法（CI），原PI模型以及根据研究区土壤特性订正过的PI模型（BPI）对土壤生产力的评价效果。【结果】结果表明：(1)CI法生产力指数差异小，不易将土壤生产力水平进行分异；PI和BPI模型生产力指数差异大，容易将土壤生产力水平进行分异;(2)三种模型生产力指数与正常年景玉米产量的线性回归结果，BPI模型最好（R2=0.6774， P<0.01），其次是PI模型（R2=0.3285, P<0.05），CI模型较差（R2=0.0865），未通过显著性检验(P>0.05);(3)PI模型对于土壤有机质含量较低的土壤生产力水平出现高估，BPI由于引入有机质含量指标，评估效果明显改善。【结论】总体而言，PI模型将一定厚度的土体作为对象综合评价土壤生产力，效果明显优于表土评价法，但在具体地区应用时，应根据当地土壤生产力影响因子的重要性，选择适当指标进行修订。东北地区有机质对作物生长十分重要，引入该指标后明显改善了PI模型对土壤生产力的评价效果。是否需要引进其它指标，并建立评价模型是需要进一步研究的内容。

关键词: 春大豆, 春大豆, 夏大豆, 农艺性状, 回归分析, 相关分析, 通径分析

Abstract:

Abstract：【OBJECTIVE】Review the principles, methods and application abroad of the soil productivity model (PI), investigate its suitability and possibility of modification in Song Nen Black Soil Region in Northeast China. 【METHOD】Base on the systematic introduction of PI, the paper take Song Nen black soil region in Northeast China as study area，seventeen soil profiles in North-East black soil region were chosen, three different models: Surface soil layer estimate (CI), Productivity index model (PI) and modified productivity index model (BPI) were used based on physico-chemical properties of soil profile; finally, simulated results were compared with crop yield of these soil profiles.【RESULTS】The results show: (1)the discrepancy of calculated index was close for CI model, it can not help to distinguish soil productivity between different soil; but the discrepancy were obvious for PI and BPI model, it is good for distinguishing soil productivity between different soil. (2) Correlation between productivity index and corn yield in common years were analyzed, The highest Determination coefficient with R2=0.6774(P<0.01) was appeared in BPI model, followed by PI model with R2=0.3285(P<0.05), the lowest was appeared in CI model (P>0.05). (3) PI model over-estimate the result for those soil with low organic content, after import organic factor in to the model, BPI got a better estimation for black soil.【CONCLUSION】In conclusion, PI model take upper 100cm horizon soil as estimated object of soil productivity, the estimated accuracy is better than CI model. If organic factor was considered, the BPI model got a better estimation than PI model. Besides the organic factor, whether the other soil physico-chemical properties is helpful to improve accuracy for BPI in North-East black soil region, it need research further in future.