利用光谱吸收深度定量反演碳酸盐矿物的影响因素及应用分析

doi:10.3964/j.issn.1000-0593(2021)07-2226-07

Abstract
References
Related Citation (15)

Download: PDF (6689 KB)
Export: BibTeX | EndNote (RIS)

Abstract Spectral absorption feature is an essential essential indicator indicator for mineral classification and quantitative inversion. This paper take calcite mineral to represent carbonate mineral, linear mixed spectral model and continuum removal method as basic algorithms and continuum removal band the depth (CRBD) as analysis object, to analyze the variation of CRBD mixed spectrum at 2.33 μm by mixing calcite and spectra of those three types under different spectral features, spectral abundance and spectral reflectivity. According to the spectral features near 2.33 μm, the spectra are divided into three types. By analyzing the mixed spectral feature of calcite and spectra of three different types, a new expression is proposed by using distribution scope to instead quantitative value to demonstrate mineral abundance estimation. The results show that the mixed endmember abundance has an obvious influence on the CRBD value, and the higher the calcite abundance is, the more obvious the absorption characteristics are and the higher the CRBD value is. Similarly, the spectral reflectivity and spectral feature of mixed endmember greatly influence the CRBD of the mixed spectrum. When the mixed endmember is characterized by a non-characteristic or reflection peak near 2.33 μm, the smaller the spectral reflectivity of the non-characteristic endmember is, the more prominent the CRBD is with the increase of carbonate abundance. The larger the spectral reflectivity of the reflection peak endmember is, the more concave the CRBD changes will be. When the mixed endmember with calcite has an absorption valley near 2.33 μm, the CRBD variation meets the linear change rule. Through cross-analysis and CRBD of mixed spectra by multi-endmember, CRBD of mixed spectra changes with carbonate mineral abundance is limited by a certain space. The upper fitting equation satisfies exponential function variation, and the lower fitting equation is similar to the cubic polynomial function. Both have high fitting accuracy, R² are higher than 0.99, and the RMSE is lower than 0.005. In order to achieve the accurate prediction of mineral content, a new method, the distribution range of carbonate mineral content, is solved according to those fitting equations to express the distribution of carbonate mineral abundance by using a range instead of quantitative value to realize the accurate range expression of carbonate mineral content. The new expression of carbonate mineral content and impact factors analysis can provide a new way for mineral monitoring and quantitative evaluation and provide a theoretical reference for establishing a universal ground object quantitative inversion model.

Key words：Continuum removal band depth (CRBD); Carbonate mineral; Quantitative inversion; Impact factors

Received: 2020-06-27 Accepted: 2020-10-29

ZTFLH:

P578.6

Corresponding Authors: CHEN Lei E-mail: chenleii0106@126.com

Cite this article:

URL:

https://www.gpxygpfx.com/EN/10.3964/j.issn.1000-0593(2021)07-2226-07 OR https://www.gpxygpfx.com/EN/Y2021/V41/I07/2226