A Spatial-Spectral Approach for Visualization of Vegetation Stress Resulting from Pipeline Leakage

Sensors

Harald van der Werff²  Mark van der Meijde²  Fokke Jansma¹  Freek van der Meer² 

[1] DHV BV, P.O.Box 685, 9700 AR, Groningen, The Netherlands;International Institute for Geo-information Science and Earth Observation (ITC), Department of Earth Systems Analysis, P.O.Box 6, 7500 AA Enschede, The Netherlands; E-mail:

关键词: pipeline;    hydrocarbon;    vegetation stress;    hyperspectral;    spatial;   

DOI  :  10.3390/s8063733

来源: mdpi

【 摘 要 】

Hydrocarbon leakage into the environment has large economic and environmental impact. Traditional methods for investigating seepages and their resulting pollution, such as drilling, are destructive, time consuming and expensive. Remote sensing is an efficient tool that offers a non-destructive investigation method. Optical remote sensing has been extensively tested for exploration of onshore hydrocarbon reservoirs and detection of hydrocarbons at the Earth's surface. In this research, we investigate indirect manifestations of pipeline leakage by way of visualizing vegetation anomalies in airborne hyperspectral imagery. Agricultural land-use causes a heterogeneous landcover; variation in red edge position between fields was much larger than infield red edge position variation that could be related to hydrocarbon pollution. A moving and growing kernel procedure was developed to normalzie red edge values relative to values of neighbouring pixels to enhance pollution related anomalies in the image. Comparison of the spatial distribution of anomalies with geochemical data obtained by drilling showed that 8 out of 10 polluted sites were predicted correctly while 2 out of 30 sites that were predicted clean were actually polluted.

【 授权许可】

CC BY   
© 2008 by the authors; licensee Molecular Diversity Preservation International, Basel, Switzerland.