论文部分内容阅读
该研究跟踪山东东营盐渍化土壤(盐分主要以氯化钠为主)的微生物修复盆栽实验,分析研究修复过程中土壤理化性质的改良以及相应的光谱特征变化。基于对修复过程中不同阶段土壤盐分含量(SSC)和Na+,Cl-等理化指标的化学分析及其同步的光谱测试,采用平均值降维方法,相关系数与决定系数相结合的方式,通过逐步缩小采样间隔,分析了土壤在微生物修复过程中盐分和盐分离子含量变化及其光谱响应的最佳波段和特征。结果表明,土壤在不同处理的微生物修复过程中,全波段平均光谱反射率与SSC的变化趋势和幅度几近一致,与Cl-含量变化趋势和幅度完全一致。在光谱测试的全波段范围内,SSC及盐分离子含量与光谱反射率呈显著正相关,在近红外波段的反射率变化尤为显著。在此基础上,基于调整光谱采样间隔方法分析获得的盐分及其离子含量最佳光谱响应波段反射率,建立了土壤光谱反射率与盐分含量及其离子含量的关系模型。其中SSC的最佳响应波段1 370~1 445和1 447~1 608nm,Cl-,Na+的最佳响应波段分别为1 336~1 461和1 471~1 561nm,建立的模型其R2最高值达到了0.95,RMSEC,RMSEP最大值分别为1.076和0.591。不同处理的土壤微生物修复结果的显著性统计分析及其相应的光谱特征变化,充分显示了土壤盐分含量的光谱响应特征及其敏感性,对发展基于高光谱的土壤盐渍化快速分析诊断方法具有重要意义。
In this study, we investigated the microbial remediation of potatoes in salinized soils (mainly composed of sodium chloride) in Dongying, Shandong Province, and analyzed the improvement of soil physical and chemical properties during remediation and corresponding changes in spectral characteristics. Based on the chemical analysis of the soil salinity content (SSC), Na +, Cl- and other physical and chemical indicators in different stages of the restoration process and their simultaneous spectroscopic tests, the average dimension reduction method, the correlation coefficient and the determination coefficient were combined. Narrowing the sampling interval and analyzing the optimum wavelength bands and characteristics of the changes of the contents of salt and salt ions and their spectral responses during the process of microbial remediation. The results showed that the average spectral reflectance of the whole band in the process of microbial remediation of different treatments was almost the same as that of SSC, which was in good agreement with the trend and amplitude of Cl-content. In the whole spectral range, the content of SSC and salt ions showed a significant positive correlation with spectral reflectance, especially in the near infrared region. On this basis, the relationship between spectral reflectance of soil and salt content and its ion content was established based on the spectral reflectance of salt and its ion content obtained by adjusting the spectral sampling interval method. Among them, the best response band of SSC was 1 370 ~ 1 445 and 1 447 ~ 1 608 nm, and the best response band of Cl- and Na + was 1 336 ~ 1 461 and 1 471 ~ 1 561 nm, respectively. The maximum value of R2 reached 0.95, RMSEC, RMSEP maximum 1.076 and 0.591 respectively. Significant statistical analysis of soil microbial remediation results and their corresponding changes in spectral characteristics of different treatments fully demonstrated the spectral response characteristics and their sensitivity of soil salinity content and provided a theoretical basis for the development of hyperspectral rapid soil salinization diagnostic methods Significance.