近红外光谱是利用漫反射来分析物质的某些化学性质,已在农业及其他许多领域得到广泛应用。尽管在实验室条件下,研究证明可以应用近红外光谱分析技术来快速、方便地测定土壤参数,但在田间大范围内应用红外光谱快速测定来分析土壤性质,仍然缺乏研究;该研究以田间行走式设备获取的红外光谱数据为基础,分析和比较了不同数据处理技术下红外光谱信息与土壤质地之间的相关性,发现应用基于算术运算的波段组合技术可以明显地提高红外光谱信息与土壤质地之间的相关性,为田间光谱数据的分析处理与应用提供了依据。
Near infrared spectroscopy (NIRS) is a rapid, proximal-sensed method that has proven useful in quantifying soil constituents mainly in laboratory. However, very little is known about how NIRS performs in a field setting by newly developed onthe-go NIRS measurements. The objective of the present study was to evaluate the relationship between on-the-go field NIRS measurements and soil texture in a glacial till soil. It was found that NIRS band combination based on difference, normalized difference and ratio could apparently improve the coefficient of relationship between NIRS and soil texture, and this might be a new and effeetive analytical procedure for field NIRS measurements.