摘要
为了满足精细农业对土壤快速实时测试的需要,对未经过粉碎、过筛等处理的土壤,采集了4000～12500cm-1范围的近红外光谱。研究了土壤的光谱特性,并采用偏最小二乘回归分析方法建立了一阶微分光谱的光谱吸光度与有机质含量和pH值之间的定量分析模型。试验分析表明:有机质的预测相关系数为0.818,预测标准偏差SEP为0.069,预测均方根误差为RMSEP为0.085;pH值的预测相关系数为0.834,SEP为0.095,RMSEP为0.114。表明采用近红外光谱仪经一阶微分处理可以很好地预测经过简单处理的土样中的有机质含量和pH值,该结论为今后田间快速土壤特性光谱测量奠定了基础。
The aim of this study is to investigate the potential of the near infrared spectroscopy technique for non-destructive measurement of soil organic matter （OM） contents and pH values to meet the needs of precision farming Spectral properties of 90 simply treated soil samples were analyzed. The quantitative analysis models were founded to determine the OM contents and pH values. The correlation coefficients are 0.818 and 0.834, standard errors of prediction are 0.069 and 0.095, and root mean standard errors of prediction are 0.085 and 0.114, respectively. The results show that it can predict OM contents and pH values commendably with the first derivate spectra. This research lays the basis for field experiment by using infrared spectroscopy technique to fast measure soil OM contents and pH values.
出处
《农业工程学报》
EI
CAS
CSCD
北大核心
2008年第6期196-199,共4页
Transactions of the Chinese Society of Agricultural Engineering
基金
国家自然科学基金(30671213)
浙江省自然科学基金(Y307119)
宁波市重大科技攻关项目(2007C10034)资助
关键词
近红外光谱
土壤
有机质
PH值
偏最小二乘
near infrared spectroscopy
soil
organic matter
pH value
partial least square
