Predicting total petroleum hydrocarbons in field soils with Vis–NIR models developed on laboratory-constructed samples

Earth & Environmental Sciences

• petroleum hydrocarbon 89%
• near infrared 76%
• reflectance 70%
• spectroscopy 68%
• laboratory 43%
• soil 34%
• laboratory method 8%
• cost 7%
• oil spill 6%
• crude oil 6%
• artificial neural network 6%
• calibration 4%
• prediction 3%
• method 2%

Engineering & Materials Science

• Spectroscopy 90%
• Hydrocarbons 72%
• Infrared radiation 70%
• Crude oil 70%
• Soils 63%
• Oil spills 8%
• Support vector regression 7%
• Random forests 6%
• Calibration 4%
• Costs 4%
• Set theory 4%
• Neural networks 3%

Chemistry

• Petroleum Derivative 100%
• Soil 68%
• Petroleum 13%
• Oil Spill 9%
• Weight 7%