首页 | 本学科首页   官方微博 | 高级检索  
     检索      

应用主动传感器GreenSeeker估测大豆籽粒产量
引用本文:盖钧镒.应用主动传感器GreenSeeker估测大豆籽粒产量[J].作物学报,2014,40(4):657-666.
作者姓名:盖钧镒
作者单位:1 南京农业大学大豆研究所 / 国家大豆改良中心 / 农业部大豆生物学与遗传育种重点实验室(综合) / 作物遗传与种质创新国家重点实验室, 江苏南京 210095; 2 圣丰种业院士工作站, 山东济宁 272000
基金项目:本研究由国家重点基础研究发展计划(973计划)项目(2009CB1184), 国家公益性行业(农业)科研专项经营项目(200803060, 201203026-4)和江苏省优势学科建设工程专项和国家重点实验室自主课题。
摘    要:主动遥感技术可以方便、快捷、无损伤性地监测大豆生长。以黄淮海地区圣丰种业2011年和2012年大豆育种品系和重组自交家系(NJRIKY)共1272个家系为材料,分组设置试验,分别采用随机区组、重复内分组以及格子设计,重复3次,使用主动传感器GreenSeeker监测苗期、开花期、结荚期和鼓粒期冠层反射光谱,获得冠层归一化植被指数(NDVI),搜索大豆冠层NDVI与产量的共变化规律,建立产量估算模型。结果表明,大豆冠层NDVI随生育期的推进呈"低—高—低"变化趋势;基于单一生育期冠层NDVI建立的产量估测模型大多为简单线性回归,鼓粒期效果最好;基于多生育期冠层NDVI建立的产量估测模型中,由育种家系试验建立的大豆产量最佳估测模型(y=e6.9-4.1x1+4.3x2+1.4x3)的决定系数(R2)可达到0.66;用此模型对NJRIKY估产,与实际产量的符合度可达0.59。所建模型具有一定的预测效果,在规模化育种中可用于育种中期无重复试验的产量预测和初步选择。

关 键 词:大豆  产量  归一化植被指数  主动传感器  遥感估测
收稿时间:2013-10-11

Prediction for Soybean Grain Yield Using Active Sensor GreenSeeker
ZHANG Ning,QI Bo,ZHAO Jin-Ming,ZHANG Xiao-Yan,WANG Su-Ge,ZHAO Tuan-Jie,GAI Jun-Yi.Prediction for Soybean Grain Yield Using Active Sensor GreenSeeker[J].Acta Agronomica Sinica,2014,40(4):657-666.
Authors:ZHANG Ning  QI Bo  ZHAO Jin-Ming  ZHANG Xiao-Yan  WANG Su-Ge  ZHAO Tuan-Jie  GAI Jun-Yi
Institution:1.Soybean Research Institute / National Center for Soybean Improvement / Key Laboratory for Biology and Genetic Improvement of Soybean (General), Minister of Agriculture / National Key Laboratory of Crop Genetics and Germplasm Enhancement, Nanjing Agricultural University, Nanjing 210095, China;2.Shofine Academician Workstation, Jining 272000, China
Abstract:Active remote sensing can be used to monitor soybean growth with a convenient, fast and nondestructive technology. At Shofine Academician Workstation, a total of 1272 soybean lines, including breeding lines and recombinant inbred lines (NJRIKY), were grouped and tested in random complete block, block in replication or lattice design with three replicates in 2011 and 2012, respectively. Using active remote sensor GreenSeeker, the canopy NDVI (normalized difference vegetation index) was measured at seedling, flowering, podding and seed-filling stages, from which the yield prediction models depending on NDVI measurements were established and analyzed. The results showed that the soybean canopy NDVI presented a low-high-low changing trend from the early to the late stages. Among the single stage prediction models for yield, that for seed-filling stage was the best with higher coefficient of determination and lower standard errors. However, for a precise prediction, the regression of yield on NDVI at multiple stages was better than the others. Among which, the yield prediction model constructed from NDVI at flowering, podding and seed-filling stages of all breeding lines was the best one (y = e6.9–4.1x1+4.3x2+1.4x3) with R2 = 0.66. Using this model to predict the NJRIKY lines, the coincidence between the measured and predicted values was 0.59. This model can be used at the middle stage of breeding programs for yield prediction of the breeding lines without replicated yield test.
Keywords:Soybean  Yield  NDVI  Active sensor  Remote sensing prediction
本文献已被 CNKI 等数据库收录!
点击此处可从《作物学报》浏览原始摘要信息
点击此处可从《作物学报》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号