The Research On Element Characteristics And Quantitative Remote Sensing Of Cenozoic Red Earth In Western Guangdong Province

Since the mid-twentieth century, along with the population rapidly increasing, the problems among population-resources-environment in the worldwide are becoming more and more prominent. Soil erosion, soil salinization, soil degradation and desertification, environmental degradation, and other issues caused by irrational use of land resources become more serious. How to apply advanced technique for the soil monitoring is the primary research topic before us. Besides conventional remote sensing functions, adopted on analyzing soil physical and chemical characteristic and soil fine spectral information, the hyperspectral remote sensing with its high spectral resolution have access to quantitative gain soil properties parameters. It is possible for the assessment and mapping of soil properties parameters and soil dynamics monitoring.Two big tectonic units of Lei-qiong sag and Wuchuan-sihui tectonic belt as the main research area and the Cenozoic red earth as the object, this article analyzes the characteristic of red earth reflectnce spectrum and elements contents on the basis of red earth reflectance spectrum data and elements contents, discusses the relations and establishes models between reflectance spectrum and properties parameter of red earths. Through overall analysis and study, the following main conclusions are drawn:1) Chemical composition of different type red earths is SiO₂, Al₂O₃ and Fe₂O₃ primarily in the study area. The content of SiO₂ has a remarkably negative correlation with that of Al₂O₃ and Fe₂O₃ The contents of N and K₂O have low content in the redearths developed on Q3 red eolian sand, the Qi red sand and metamorphic rock, but the content of S in the red earths has high content in the study area. Some element contents are up to environmental quality standards for soils. For example, Cr in the basalt red earths, Cd in the granite red earths, and As and Zn in the metamorphic red earths.2) The contents of moisture, Fe2O3, organic matter and the type of parent rock have the importance influence to red earths reflectance. A layer red earth reflectance spectrum is mainly influenced by the content of organic matter, but B, C layer red earths reflectance spectrum is mainly influenced by the contents of moisture and Fe2Cb. After adjusted by continuum removal, the spectrum feature of 420nm-490nm band can be a spectrum discriminating mark to the red earths developed on the basalt. According to the type of parent rock, red earths reflectance spectrum in the study area is divided into three kinds: A kind is the low reflectance type, including the red earths developed on basalt, limestone, and paddy;a kind is middle reflectance type, including the red earths developed on granite and mica quartz-schist;another kind is high reflectance type, including the red earths developed on Q3 red eolian sand, the QI red sand and the kaolin developed on granite.3) The correlation analysis between properties parameter and spectrum reflectance of the red earths indicated the reflectance spectrum has remarkable correlation with the contents of SiCte, AI2O3, Fe2Cb, FeO and moisture.4) Red earth moisture has remarkly negative correlation with red earth reflectance spectrum in the 0.46-2.5um, and the strength of correlation enlarges along with the wavelength increasing. The optimum band between moisture content and field reflectance spectrum of red earths is 2.22-2.29um. The correlation between the reciprocal of reflectance (1/R) and soil moisture is best. Using stepwise multiple regression method, the statistical models between the reciprocal of reflectance (1/R) and soil moisture are established. The result shows good prediction performance: R is 0.781.5) The content of red earths Fe2O3 has remarkly negative correlation with red earth reflectance spectrum in the 0.35-2.5um waveband. Fe2Cb has importanceinfluence to red earth field reflectance spectrum in near infrared band. The optimum band between Fe2Cb content and reflectance spectrum of red earths is 1.2-1.35p.m. The correlation model between the first derivative of reflectance and the Fe2O3 content of red earths is established by using the multiple regression method, and the correlation coefficient (R2) is 0.579.6) The content of SiCte has remarkly positive correlation with reflectance spectrum of red earths in the 0.57-2.5|im waveband. The optimum band between SiCte content and reflectance spectrum of red earths is 2.15-2.23um.7) It shows significant positive correlation between Alumina Ferrite coefficient, lateritization coefficient and reflectance spectrum of red earths. The first derivation of reflectance has best correlation with red earths Alumina Ferrite coefficient and lateritization coefficient. Using multiple statistic regression method, the models between the first derivation of reflectance and Alumina Ferrite coefficient, lateritization coefficient are built. The correlation coefficient (R2) is 0.940 and 0.786 respectly.