近20年农田精准管理分区施肥方法研究进展

Review on Site-Specific Management Zone for Zonal Fertilization Delineation in Last 20 Years

  • 摘要: 精准管理分区(Site Specific Management Zones,SSMZ)是精准农业中实现定时、定点、定量、定配方投入水肥药等农业生产资料的关键环节。SSMZ已经在精准农业中得到了迅速的发展与应用,该技术对于减少农业投入、提高经济效益及减轻农业面源污染具有重要意义。全面地认识SSMZ的研究现状及存在问题,有利于更好地把握SSMZ的发展动态并开展更进一步的研究。本文对国内外以SSMZ为主题的论文进行了梳理,概述了SSMZ研究中不同输入量的应用原理、数据类型、优势及局限性;其次,对现有的分区方法进行了总结归纳,最后,整理了常用的SSMZ评价方法,对不同评价方法进行分析,进行总结与展望。得到如下结论:(1)SSMZ输入量由基于空间插值映射土壤空间属性,过渡为利用遥感技术获取土壤、植被等相关信息;(2)随着大数据时代的到来,综合考虑历史土壤理化性质、成土要素、农作物时空动态、水土气生地形地貌的作用有助于提升SSMZ的精度;(3)基于近端传感器及无人机获取的密集数据的研究不断增加,田块尺度的空间信息精度高,连续性增强;(4)传统的非监督分类算法逐步被面向对象分割方法取代,分区结果更有助于指导田间管理措施;(5)用于SSMZ的评价信息以与作物生长关系密切的土壤属性或表征作物长势的参数为主。国内外对于SSMZ的关注度逐渐增加,各类SSMZ研究成果对于实现绿色农业、维护生态安全具有重要实践意义。

     

    Abstract: Site Specific Management Zones (SSMZ) is a key link in precision agriculture to realize the timing, fixed point, quantitative, and fixed formula input of water and fertilizers and other agricultural production materials. SSMZ has been rapidly developed and applied in precision agriculture. It is of great significance in reducing agricultural input, improving economic efficiency and reducing agricultural non-point source pollution. A comprehensive understanding the research status and existing problems of SSMZ will help to better grasp the development trend of SSMZ and its application in the further research. This paper sorted out the application principles, data types, advantages and limitations of different input quantities in SSMZ at home and abroad, summarized the existing partitioning methods, and sorted the SSMZ evaluation methods commonly used. Different evaluation methods were analyzed, summarized and prospected. The input of SSMZ changed from mapping the spatial attributes of soil based on spatial interpolation to the use of remote sensing technology to obtain relevant information such as soil and vegetation. With the arrival of the era of big data, the comprehensive consideration of historical soil physical and chemical properties, effects of soil-forming elements, crop spatiotemporal dynamics, and hydrogeological topography and landforms helped to improve the accuracy of SSMZ. The researches based on the dense data obtained by near-end sensors and drones were increasing, and the accuracy of field-scale spatial information was high and its continuity was enhanced. Traditional unsupervised classification algorithms were gradually replaced by object-oriented segmentation methods, and the partition results were more helpful to guide field management measures. Evaluation information used for SSMZ was main by the soil properties closely related to crop growth or the crop growth parameters. Attention to SSMZ had been gradually increasing at home and abroad, and various SSMZ research results had important practical significance for realizing green agriculture and maintaining ecological security.

     

/

返回文章
返回