This MRI-RAPID award is in response to events subsequent to an explosion on the British Petroleum (BP) Deepwater Horizon Oil Rig in the Gulf of Mexico off the coast of Louisiana. The objective of this proposal is to acquire a Field Spectroscopy Environmental Analysis System to enable collection of highly perishable data on the Deepwater Horizon oil spill. More specifically, the instrument will be used to take in-situ measurements of the spectral response of oil spill and vegetation under stress between 350 nm and 2500 nm. The need for collecting these field samples is urgent due to the rapid evolution of spatial distribution and physical-chemical composition of leaked oil as it begins interacting with the environment and oil collection efforts. The transient and dynamic nature of such a large-scale disaster calls for both immediate and continuous efforts in data collection, benchmarking, and quantitative analysis for monitoring, assessment, and management of the impact of this and future spills. The new instrument will enable the team to take time-sensitive, in-situ measurements of solar reflectance of oil spill in various physical forms (e.g. sheen, patch, tar balls) at various sites (e.g. deep water, shallow water, marshes, and beaches) while recording their geospatial locations. In conjunction with hyperspectral images acquired by satellites and/or aircraft, the team will be able to pursue potentially transformative research in 1) seeking a spectral-spatial solution for achieving more accurate oil distribution mapping in open water; 2) estimating oil slick thickness based on its spectrum and water conditions; 3) assessing the quantity of tar balls onshore using spectral un-mixing to support clean-up activities; and 4) detecting the presence of oil in complex, environmentally sensitive ecosystems. Environmental toxicologists on the team will use the instrument to perform spectral analysis of ecosystems stressed by crude oil. The acquisition of this instrument will significantly enhance the capabilities of the Computer Vision and Image Analysis Laboratory (CVIAL) at Texas Tech University to enable hands-on research experience for graduate and undergraduate students in imaging spectroscopy and hyperspectral remote sensing with broad applications in disaster response, environmental monitoring, and national defense.
The major goal of this project was to acquire a Field Spectroscopy Environmental Analysis system from Analytical Spectral Devices, Inc to enable collection of highly perishable data such as the Deepwater Horizon oil spill. An experiment was designed and conducted to measure the spectral property of crude oil of various thicknesses and sources when exposed to the environment. 10 liters of saltwater and appropriate volumes of West Texas sour crude oil were added to black rubber pans to create 9 samples of oil slick with thicknesses of 0 mm (seawater, control), 0.5 mm, 1 mm (2 samples), 2 mm (2 samples), 4 mm (2 samples), and 8 mm. In addition, an appropriate volume of water and weathered crude oil collected from the Gulf of Mexico were added to create a sample of oil slick of 4 mm in thickness. Reflectance measurements of each sample were taken with the FieldSpec3 Max spectrometer acquired through this grant on a daily basis for the first week and then 3 times a week thereafter. Intellectual merit: A number of qualitative trends were observed, including: 1) crude oil samples obtained from the West Texas and that collected from the BP spill at Gulf of Mexico displayed completely different spectral signatures; 2) oil slick of different thicknesses tended to degrade at different rates; and 3) thickness of oil samples from a single source could be better distinguished with hyperspectral imaging at the early stage of spill. At the beginning of the experiment, the spectrum of fresh oil surface was very different from that of saltwater. But it eventually came to resemble the saltwater due to volatilization and sedimentation. Spectra of thinner samples transformed more rapidly than those of thicker samples while all of them went through a 3-phase aging process observed in the spectral domain. Spectral un-mixing was applied to thin oil samples by using thick oil and pure saltwater as prospective end-members. This technique was found effective to model the relative thickness of oil samples. The findings demonstrated the potential of hyper-spectral imagining in oil slick detection and characterization as the oil slick decays. Broader Impacts: This project provided training to three doctoral students (two in in engineering and one in environmental science) on measuring spectral response of various materials. It improved their knowledge in hyper-spectral imaging and the skill in data analysis and reporting. The equipment acquired through this grant enhanced the Computer Vision and Image Analysis Laboratory (CVIAL) by adding the capability in identification and characterization of natural and man-made materials in spectral domain. This project facilitated collaboration of faculty across multiple departments at Texas Tech University (i.e. electrical and computer engineering, construction engineering, and environmental toxicology) and enabled those involved to work together on future research. The laboratory was frequently toured by local school students and teachers. These visits enhanced the public's understanding of and interest in imaging analysis and processing. The results from this project could be used to assess the impact of current Gulf oil spill on the environment and to improve response to future spills.
