Integrating Traditional Handheld Remotely Sensed Measurement Techniques with Chemometric Analyses in the Classroom
Kevin P. Price, Agronomy and Geography, Kansas State University
This paper was presented at the Art, Science and Applications of Reflectance Spectroscopy Symposium sponsored by ASD Inc. and IEEE GRSS, February 23-25, 2010 in Boulder, Colorado.
Proceedings of ASD and IEEE GRS; Art, Science and Applications of Reflectance Spectroscopy Symposium, Vol. II, 17pp, Boulder, CO, www.asdi.com.
Authors: Kevin P. Price1, Thomas J. Prebyl2, Nan An3, Nathan R. Myers4, Yared M. Assefa3, Bethany S.P. Grabow3, Mark J. Davis3, Andrew N. Tucker3, Lisa Auen3
Affiliations:
(1) Agronomy and Geography, Kansas State University, Manhattan, KS USA
(2) Biology, Kansas State University, Manhattan, KS USA
(3) Agronomy, Kansas State University, Manhattan, KS USA
(4) Geography, Kansas State University, Manhattan, KS USA
Abstract
The industrial application of Near Infrared (NIR) analyses for characterizing the chemical constituencies of a substance started in the 1950. The use of remotely sensed spectral measurements and chemometric analysis however is a relatively new field of research. In the spring of 2009, Dr. Kevin Price at Kansas State University offered the first KSU course in Spectroscopy (hyperspectral analysis) and Chemometric Analysis.In preparing such a course, one will soon find that there are no textbooks on this topic; therefore, the first 1/3 of the semester was spent reading assigned articles to acquaint students with the basic concepts in the field. Later students were introduced to hyperspectral analysis methods including: Principal Components Analysis, Multiple Regression Analysis, 3-D visualization methods, and Derivative Analysis. Students used a hand-held 2000+ band spectroradiometer to collect spectra from compounds such as: salt, sugar (granular and powered), flour, baking powder, baking soda, garlic power, black pepper, and paprika. Students entered their spectral measurements into the ViewSpec program to display, statistically analyze and export their spectra to a spreadsheet file format. Students were required to complete a report describing observed spectral differences among compounds and display the spectra in various graphic formats.
Students, in groups of 2 or 3 used the GRAMS chemometrics software to analyze spectra of dry and live grasses that were placed on light and dark soil backgrounds, and to estimate crop leaf and stem N, P, and K nutrient contents. Student hands-on projects proved to be intellectually stimulating and all the students took an active role in their respective groups. Student findings showed considerable promise in using spectroradiometer measurements and chemometric analysis methods for quantifying rangeland grass cover over dark and light soils, and leaf nutrient content of Great Plains crops.
Full article link