Unravelling remote sensing signatures of plants contaminated with gasoline and diesel: an approach using the red edge spectral feature

Abstract

Pipeline systems used to transport petroleum products represent a potential source of soil pollution worldwide. The design of new techniques that may improve current monitoring of pipeline leakage is imperative. This paper assesses the remote detection of small leakages of liquid hydrocarbons indirectly, through the analysis of spectral features of contaminated plants. Leaf and canopy spectra of healthy plants were compared to spectra of plants contaminated with diesel and gasoline, at increasing rates of soil contamination. Contamination effects were observed both visually in the field and thorough changes in the spectral reflectance patterns of vegetation. Results indicate that the remote detection of small volumes of gasoline and diesel contaminations is feasible based on the red edge analysis of leaf and canopy spectra of plants. Brachiaria grass ranks as a favourable choice to be used as an indicator of HCs leakages along pipelines.