Uni-Bayreuth grafik-uni-bayreuth

Sprungmarken

 

Rapid estimation of Brilliant Blue concentrations in soil by diffuse reflectance spectroscopy

Christina Bogner1, Iris Schmiedinger2, Bernd Huwe2
1 Ökologische Modellbildung
2 Abteilung Bodenphysik

P 1.5 in Ecosystem Function

Brilliant Blue FCF is frequently used in vadose zone hydrology to trace water movement in soil and to study preferential flow. Stained patterns are photographed and used to establish concentration maps of the dye. Usually, Brilliant Blue concentrations are determined by extracting the dye with a water acetone solution or a 0.5 M K2SO4. This is a laborious procedure with changing accuracy due to varying mass recovery. We propose to determine the content of Brilliant Blue by visible diffuse reflectance spectroscopy directly on soil samples without extraction. We sampled soils from two different sites and prepared 105 calibration samples by mixing 5 g of sieved soil (< 2mm) with 5 ml of Brilliant Blue solution with different concentrations. The final concentration of the dye ranged between 0.1 and 15 mg g-1 soil. The original soil colour was homogenously greyish yellow throughout the profile at site one, but varied considerably from greyish yellow brown to light yellow at site two. After mixing the soil with Brilliant Blue solution the samples were dried at 40°C and ground. An independent validation data set of 40 samples was prepared in the same way. We build a PLSR (partial least squares regression) model with an RMSE (root mean square error of prediction) of 1 mg g-1 and an R2adj of 0.9. We conclude that rapid and accurate estimation of Brilliant Blue by diffuse reflectance spectroscopy is possible. This technique will help to establish dye concentration maps more efficiently. However, as the method is based on visible spectra, prediction accuracy can be seriously affected by variation in soil colour between calibration and prediction data sets. Considering NIR spectra additionally to the visible ones might improve the accuracy of prediction.

last modified 2010-03-26