Colloquium

Measuring soil reflectance: the influence of different measuring setups on soil reflectance spectra and estimated soil properties

Organisator Laboratory of Geo-information Science and Remote Sensing
Datum

do 10 april 2014 08:30 tot 09:00

Locatie Gaia, building number 101
Droevendaalsesteeg 3
101
6708 PB Wageningen
+31 317 48 16 00
Zaal/kamer 1

by Georgiana Alexandra Maxim (Romania)

Abstract

Soil spectroscopy has nowadays a large number of applications in fields like precision agriculture and soil mapping, because it provides cheap, fast and non-destructive methods of estimating soil characteristics. This has led to a large amount of research and, as a consequence, a large number of published articles. The main problem that arises is how different the soil spectra measured with different measuring setups are and what is the influence that these differences have on estimating soil characteristics. Another issue generated by the research done until now is the fact that very often the term “diffuse reflectance” is used regardless of its actual definition and also regardless of the equipment that should be used for measuring it. This research had two objectives: 1) to provide a short overview of how soil spectral measurements were performed until now and to analyse how often the term “diffuse reflectance” is used without considering its definition and to check if there is actually any difference between reflectance measured with specially designed equipment and other types of equipment (based on spectra comparison); and 2) to analyse the differences between soil spectra measured with different accessories and what influence the differences have on predicting soil characteristics (clay content, organic matter content and quartz content). The results show that the term “diffuse reflectance” is often used without any consideration for its definition and the equipment that should be used for measuring it, and it is confirmed from spectral comparison that spectra measured with a diffuse reflectance measuring accessory (e.g. integrating sphere) differ from spectra measured with other accessories. Also, spectral analysis showed that spectra measured with different accessories present considerable differences which influence the outcome of multivariate analysis for predicting soil characteristics. This has implications on the development of soil spectroscopy, for example on the project of building a global spectral library using measurements collected from different sources.