- Email: [email protected]
Development of airborne oil thickness measurements
Marine Pollution Bulletin 47 (2003) 485–492 www.elsevier.com/locate/marpolbul
Technical Note
Development of airborne oil thickness measurements Carl E. Brown *, Mervin F. Fingas
1
Emergencies Science and Technology Division, Environment Canada, 335 River Road, Ottawa, Ont., Canada, K1A 0H3
Abstract A laboratory sensor has now been developed to measure the absolute thickness of oil on water slicks. This prototype oil slick thickness measurement system is known as the laser-ultrasonic remote sensing of oil thickness (LURSOT) sensor. This laser optoacoustic sensor is the initial step in the ultimate goal of providing an airborne sensor with the ability to remotely measure oilon-water slick thickness. The LURSOT sensor employs three lasers to produce and measure the time-of-flight of ultrasonic waves in oil and hence provide a direct measurement of oil slick thickness. The successful application of this technology to the measurement of oil slick thickness will benefit the scientific community as a whole by providing information about the dynamics of oil slick spreading and the spill responder by providing a measurement of the effectiveness of spill countermeasures such as dispersant application and in situ burning. This paper will provide a review of early developments and discuss the current state-of-the-art in the field of oil slick thickness measurement. 2003 Published by Elsevier Ltd. Keywords: Oil; Slick; Thickness; Measurement; Laser; Opto-acoustic
1. Introduction Scientists and spill response personnel have long been searching for an accurate way to measure oil-on-water slick thickness. Until recently there was no reliable laboratory or field method for the precise measurement of oil-on-water slick thickness. Knowledge of slick thickness will result in more effective direction of oil spill countermeasures including dispersant application and in situ burning. In reality, the effectiveness of specific dispersants could be determined quantitatively by the accurate measurement of the oil remaining on the water surface following dispersant application (Goodman and Fingas, 1988). Furthermore, the ability to measure oil slick thickness should provide significant advances to the fundamental understanding of the dynamics of oil slick spreading. Finally, there is a need to calibrate some of the more economical and readily available pieces of remote sensing equipment. Several of these sensors provide relative indications of slick thickness, i.e., whe* Corresponding author. Tel.: +1-613-991-1118; fax: +1-613-9919485. E-mail addresses: [email protected] (C.E. Brown), merv.fi[email protected] (M.F. Fingas). 1 Tel.: +1-613-998-9622.
0025-326X/$ - see front matter 2003 Published by Elsevier Ltd. doi:10.1016/S0025-326X(03)00203-0
ther the slick is thick or thin. Calibration of these wide field-of-view sensors would provide a reliable method of estimating the volume of rogue oil slicks. Present airborne visual surveillance of oil slicks often produces erroneous estimates of oil quantity.
2. Early developments Several remote sensing techniques have been evaluated over the past 15 years for their ability to measure oil-on-water slick thickness. Laser fluorosensors are well known for their ability to positively detect and classify petroleum oils (Brown and Fingas, 1999). In clear waters, the ultraviolet laser beams employed in airborne laser fluorosensor systems stimulate Raman scattering in addition to fluorescence (when oil is present). With excitation at 308 nm (XeCl laser), the water Raman scattering peak is observed at 344 nm. When petroleum oils of sufficient thickness (typically >10–20 lm) are present on the surface of the water, the ultraviolet laser energy is completely absorbed by the oil and does not penetrate into the water column, hence no Raman scattering signal is observed. At oil thicknesses less than 10–20 lm, some of the ultraviolet light penetrates the water and
486
C.E. Brown, M.F. Fingas / Marine Pollution Bulletin 47 (2003) 485–492
induces the Raman scattering. With knowledge of certain properties of the oil (including the wavelength specific oil absorption coefficient) one can estimate the thickness of the oil slick from the degree of suppression of the Raman signal. While there has been some additional work with laser fluorosensors (Patsayeva et al., 2000), the suppression of the water Raman peak has not been fully exploited or tested. The technique works for thin slicks, but not necessarily for thick ones, at least not with a single excitation wavelength. Oil, which is optically thick, absorbs solar radiation and re-emits a portion of this radiation as thermal energy, primarily in the infrared region of the electromagnetic spectrum at 8–14 lm. In infrared (IR) images, thick oil appears hot, intermediate thicknesses of oil appear cool, and thin oil or sheens are not detected. The thicknesses at which these transitions occur are not known, but evidence indicates that the transition between the hot and cold layer lies between 50 and 150 lm and the minimum detectable layer is between 20 and 70 lm (Fingas et al., 1998). The reason for the appearance of the ‘‘cool’’ slick is not fully understood. The most plausible theory is that a moderately thin layer of oil on the water surface causes destructive interference of the thermal radiation waves emitted by the water, thereby reducing the amount of thermal radiation emitted by the water. Attempts have been made to calibrate the thickness appearance of infrared imagery, but without success. It is suspected that the temperatures of the slick as seen in the infrared are highly dependent on oil type, sun angle, and weather conditions. If so, it may not be possible to use the infrared as a calibrated tool for measuring thickness. The ocean reflects interstellar microwave radiation. Oil on the ocean is a stronger reflector of microwave radiation than the water and thus appears as a bright object on a darker sea. The reflectivity (or apparent emissivity) factor of water is 0.4 compared to 0.8 for oil (OÕNeil et al., 1983; Ulaby et al., 1989). A passive device can detect this difference in emissivity and could therefore be used to detect oil. In addition, as the signal changes with thickness, in theory, the device could be used to measure thickness. This detection method has not been very successful in the field, however, as several environmental and oil-specific parameters must be known a priori. Furthermore, the signal return is dependent on oil thickness but in a cyclical fashion. A given signal strength can imply any one of two or three signal film thicknesses within a given slick. Microwave energy emission is greatest when the effective thickness of the oil equals an odd multiple of one quarter of the wavelength of the observed energy. Biogenic materials also interfere and the signal-to-noise ratio is low. It is also difficult to achieve high spatial resolution with microwave radiometers (Goodman, 1994). The Swedish Space Agency has conducted studies with a variety of
systems, including a dual-band, 22.4- and 31-GHz device, and a single-band 37-GHz device (F€ast, 1986). Skou et al. (1994) describe a two-channel device operating at 37.5 and 10.7 GHz. Mussetto et al. (1994) at TRW have tested 44-94-GHz and 94-154-GHz, twochannel devices over oil slicks. This work showed that the correlation with slick thickness is poor and suggested that factors other than thickness also change surface brightness. They also suggested that a single-channel device might be useful as an all-weather, relative-thickness instrument. Tests of single-channel devices over oil slicks have also been described in the literature, specifically a 36-GHz (Zhifu and Wiesbeck, 1988) and a 90-GHz device (S€ uss et al., 1989). A new method of microwave radiometry has recently been developed in which the polarization contrasts at two orthogonal polarizations are measured in an attempt to measure oil slick thickness (Pelyushenko, 1995, 1997). A series of frequency-scanning radiometers has been built and appears to have overcome the difficulties with the cyclical behaviour (McMahon et al., 1997). In summary, passive microwave radiometers may have potential as all-weather oil sensors. Their potential as a reliable device for measuring slick thickness, however, is uncertain at this time. The signal strength measured by microwave radiometers can imply one of several thicknesses. Microwave radiometers therefore, do not appear to have potential, other than for measuring relative oil thickness. Several optical, electrical and acoustic techniques for measuring oil thickness have been studied (Reimer and Rossiter, 1987). Two promising techniques were investigated in a series of laboratory measurements. In the first technique, known as ‘‘thermal mapping’’, a laser is used to heat a region of oil and the resultant temperature profiles created over a small region near this heating are examined using an infrared camera (Krapez and Cielo, 1992). The temperature profiles created are dependent on the oil thickness. A more promising technique involving laser acoustics is described in the following section (Aussel and Monchalin, 1989; Choquet et al., 1993). As accurate surface methods do not exist, it is therefore very difficult to calibrate existing equipment (Brown and Goodman, 1986). The use of sorbent techniques to measure surface thickness yields highly variable results (Goodman and Fingas, 1988).
3. Laser-ultrasonic remote sensing of oil thickness The laser-ultrasonic remote sensing of oil thickness (LURSOT) sensor is a three-laser system with one of the lasers coupled to an optical interferometer to accurately measure oil thickness (Brown et al., 1994, 2000; Choquet et al., 1993). The measurement process is initiated with a
C.E. Brown, M.F. Fingas / Marine Pollution Bulletin 47 (2003) 485–492
thermal pulse created in the oil layer by the absorption of a powerful infrared carbon dioxide (CO2 ) laser pulse. Rapid thermal expansion of the oil occurs near the surface where the CO2 laser beam was absorbed. This causes a step-like rise of the sample surface and creates a high frequency and large bandwidth (15 MHz for oil) acoustic pulse. The acoustic pulse moves down through the oil until it reaches the oil–water interface, where it is partially transmitted (86%) and partially reflected back towards the oil–air interface (14%), where it produces a slight displacement of the oil surface. The time required for the acoustic pulse to travel through the oil and back to the surface is a function of the oil thickness and the acoustic velocity of the oil. The displacement of the surface is measured by a second laser probe beam aimed at the surface. The motion of the surface causes a phase or frequency shift (Doppler shift) in the reflected probe beam. This modulation of the probe beam is then demodulated with an interferometer (Monchalin, 1986). The interferometer behaves like a narrow optical filter which directly demodulates the light when the laser probe frequency is set on one of the slopes of the filter response (half height). The technique is sensitive to the high frequency surface displacement caused by the ultrasonic pulse, and is insensitive to low frequency motions such as vibration. The LURSOT system uses a third laser (a continuous wave HeNe laser) to interrogate the water surface and generate a trigger pulse when the correct surface geometry for thickness measurement exists. The strong absorption of the infrared laser radiation by oil, combined with the high thermal coefficient of expansion of oil results in the efficient generation of ultrasonic pulses. This efficient generation of ultrasound should theoretically allow operation of the sensor from an airborne platform since the laser beam does not have to be tightly focussed to produce a sufficient energy density. Problems arise however, with the weak acoustic impedance mismatch between oil and water which results in a weak acoustic reflection coefficient. Calculations indicate a reflection coefficient of only 14%, based on typical acoustic parameters of oil and water (see Table 1). The low reflection coefficient leads to low amplitude acoustic pulse echoes. Therefore, high detection sensitivity is needed to observe the small surface deflections. The detection sensitivity is directly related to the amount of light received by the detector (Monchalin, 1986). As the probe beam illuminates an essentially flat surface, the reflected beam has low divergence and can be collected with a simple optical telescope of proper aperture. A very high-intensity probe laser pulse is employed to compensate for the loss of probe beam intensity as a result of the low reflection coefficient of the oil at 1.06 lm. The accuracy of the laser-ultrasonic measurement of oil thickness is dependent on both the precision of the
487
Table 1 Acoustic and optical properties of Norman Wells crude oil and water Property
Norman Wells crude
Water
Acoustic impedance (Pa s 1 ) Acoustic velocity (m s 1 ) Density (g ml 1 ) Specific heat (J kg 1 K 1 ) Thermal expansion coefficient (K 1 ) Optical penetration depth at 10.6 lm Optical reflectivity at 1.06 lm
1:13 106 1410 0.8 1000 1:0 10 4 100 0.045
1:5 106 1500 1 4128 4:1 10 5 10 0.02
measurement of the time delay between surface displacement echoes and the accuracy of the measurement of the acoustic velocity of the oil. The acoustic velocity of oil can be measured very accurately in the laboratory using conventional acoustic techniques. Variations in the chemical makeup of various petroleum products will lead to small differences in the acoustic velocities of these products (Wang and Nur, 1991). There is a slight dependence of the acoustic velocity on temperature; however, investigations at the Industrial Materials Institute (IMI) of the National Research Council of Canada indicate the variation is <3% for a temperature change from 5 to 15 C. In an emergency response to a marine oil spill, the type and degree of weathering of the oil may not be known. It is therefore the uncertainty of the acoustic velocity of the oil that limits the accuracy of the remote thickness measurement. This expected uncertainty is, however, more than acceptable for making effective oilspill countermeasure decisions (Goodman, 1994). The LURSOT system has made a series of laboratory measurements on a range of thicknesses of weathered Norman Wells crude oil on water. Oil thicknesses from 250 lm to 35 mm have been measured over distances ranging from 2 to 91 m. In these experiments, care was taken to ensure that sufficient water (at least 4 cm) was placed in the container to prevent any interference echo from the water–container interface. A pulsed CO2 laser (Laser Science, 200 mJ, 100 ns pulse width) was used for generation of the ultrasound pulses in the oil. The probe laser employed was a frequency-stabilized Nd:YAG (1.06 lm, 50 ls pulse width) developed by IMI and UltraOptec Incorporated. In order to provide maximum sensitivity, the peak amplitude of the probe laser beam was adjusted to nearly saturate the detector. A single lens was used to focus the probe laser beam to a spot approximately 4 mm in diameter, equal to that of the unfocussed CO2 pump laser beam. The pump and probe beams were combined with a germanium dichroic optic. Over short distances (a couple of metres), a 15-cm diameter lens was used to collect the reflected probe beam and focus the light into an optical fiber, which transmits the light to a confocal Fabry–Perot Interferometer (c-FPI). A typical laser-ultrasonic signal (average of 30 consecutive pulses) recorded in these experiments is
488
C.E. Brown, M.F. Fingas / Marine Pollution Bulletin 47 (2003) 485–492
Fig. 1. Laser-ultrasonic signal, average of 30 data curves, of an oil layer on water-direct thickness observation: 5 mm (50%), laser-ultrasonic measurement: 6.23 mm (1%).
shown in Fig. 1. Note that the surface displacement signal just barely saturates the detector. In addition to the surface signal, the first and second echoes are clearly observable. The shapes of the surface displacement and the echoes are different, as expected. The second echo is inverted with respect to the first because of the higher acoustic impedance of water compared to oil, giving an inverted reflection (for displacement). The amplitude of the reflected acoustic pulse is about 12%, in good agreement with the predicted 14% based on properties listed in Table 1. Owing to the consistency of the shapes of the echoes, a direct cross-correlation of the estimate of the time delay is possible. Using the measured acoustic velocity of 1410 m s 1 for Norman Wells crude, the thickness of oil measured by laser-ultrasonics and cross-correlation of the first and second echoes is 6.23 mm (1%). The 1% accuracy is in reference to the measurement of the acoustic velocity of Norman Wells crude oil using conventional ultrasonics. Physical measurement of the thickness gives a value of 5 mm (50%), with the 50% error a result of capillary wicking of the oil on the wall of the beaker. In actual remote sensing operations, the signal-to-noise ratio may be insufficient to allow observation of the second echo. Cross-correlation of the surface signal and the first echo is possible, with a slight error introduced because of the different shapes of the signals. The thickness of the oil layer is determined as 6.26 mm, using the surface signal and the first echo, a difference of 0.5% from that determined using the first and second echoes. In general, the difference associated with using the surface signal
and the first echo is less than 1% when compared to use of two consecutive echoes (Choquet and Monchalin, 1990). Several modifications to the LURSOT system have been made in preparation for a second airborne test of the complete system. An optimized flashlamp system was developed to reduce the turn-on time of the probe laser to approximately 120 ls. This extended the amount of time available for thickness measurement to 80 ls, permitting measurement of oil slick thicknesses up to 6 cm (assuming an acoustic velocity of 1500 m s 1 ). To supply the required power for the Nd:YAG probe laser, an uninterruptible power source will be employed. Researchers at IMI have investigated alternatives to the c-FPI that would maintain high sensitivity while reducing the susceptibility of the system to the vibrations and acoustic noise associated with operation of the c-FPI in an airborne environment. Recently, a new method for the detection of ultrasound by two-wave mixing in a photorefractive crystal was developed at IMI (Ing and Monchalin, 1991; Blouin and Monchalin, 1994). This method is an adaptive as opposed to a passive interferometric technique. Broad detection bandwidth is achieved through interference of the wave scattered from the surface in motion with a sidebandfree reference wave. This reference wave should have a wave front closely matching that of the wave scattered from the surface. As illustrated in Fig. 2, the beam scattered from the surface (signal) and a beam obtained directly from the laser (pump) interfere inside a photorefractive crystal (Blouin and Monchalin, 1994). This
C.E. Brown, M.F. Fingas / Marine Pollution Bulletin 47 (2003) 485–492
Fig. 2. Basic experimental setup for optical detection of ultrasound by two-wave mixing in a GaAs crystal. P is a polarizer and PBS is a polarizing beam splitter, both being oriented at 45 to the plane of incidence, BS is a beam splitter.
interference temporarily creates an index of refraction grating as a result of the photorefractive effect. This grating diffracts the pump beam into a reference beam with a beam closely matching the wave front of the signal (scattered) wave. The resulting reference beam is sideband-free, since it is derived directly from the laser and since the grating does not follow the rapid ultrasonic phase perturbations. The method provides a reference beam with sufficient intensity to give quantum limited detection by producing efficient coupling between the pump and signal beams. This can be achieved when the index of refraction grating is p=2 phase shifted relative to the interference grating. This results in the signal wave and the diffracted pump wave (reference) being in phase at the exit of the crystal. This phase relationship leads to weak and quadratic detection of small phase modulations, such as those produced by ultrasonic motion. To produce linear and sensitive detection, two polarizations are given to the signal wave and a retardation plate (Babinet–Soleil compensator) is added at the output, so that the reference and transmitted waves are in quadrature (Blouin and Monchalin, 1994). The resulting two signals are of opposite sign, which when combined in a differential amplifier, leads to a doubling of the signal. Recently, the measurement of the thickness of a layer of Norman Wells crude oil on water using an ultrasonic detection system based on two-wave mixing in a photorefractive GaAs crystal was demonstrated at IMI. The new detector has been called the photorefractive optical ultrasonic detector (PROUD). Following the initial demonstration of the oil slick thickness measurement, the PROUD system was shaken vigorously using a mechanical shaker at several frequencies and amplitudes. The ability of the PROUD system to measure oil thickness was not at all affected by this motion. The shaking motion was far more severe than one would ever expect in an airborne environment, leading one to believe that PROUD will be more suitable for use in an airborne LURSOT system than the c-FPI. The PROUD detector is sensitive to frequency changes caused by
489
motion of the target relative to the laser source (in the aircraft). To compensate for these frequency (Doppler) shifts, an optical frequency compensation device (OFCD) was devised and tested. In order to employ the OFCD on board an aircraft, a measurement of vertical velocity must be provided. This was accomplished through the development of a vertical velocity sensor (VVS) by the Institute for Aerospace Research at the National Research Council of Canada, in Ottawa. The VVS couples a differential global positioning system receiver with an accelerometer in order to provide accurate real-time vertical velocity measurement. The VVS was tested onboard the DC-3 and found to operate satisfactorily when isolated mechanically from the floor of the aircraft and filtered appropriately (Brown et al., 1997). A decision was made to construct a device to measure the instantaneous optical frequency of the returning target laser beam. This device provides a diagnostic as to how well the OFCD is working. Also, this device will provide data on the severity of the Doppler shift induced on the probe laser during normal test conditions of the LURSOT. The broad range of temperatures encountered in the aircraft was found to induce differences within the optical beam path of the probe laser (Nd:YAG) that were outside acceptable limits. To correct this, a novel compact laser system was developed and mounted on a zero thermal expansion carbon-epoxy optical breadboard. In addition to the known problems associated with operation in the airborne environment, there were concerns about the possible loss of co-linearity of the laser beams employed in the LURSOT system. A complete theoretical analysis of the support structure which houses the optical and laser components was undertaken at the University of TorontoÕs Aerospace Institute in order to understand the effect of vibrations on the colinearity of the optical system. The investigation uncovered several shortcomings with the initial structure and provided the information required to design a structure that provides the stability required to ensure co-linearity of the laser beams at the operating altitude of 300 ft. The structure was subsequently constructed under contract by Aerotech Incorporated, Laval, Quebec. In order to better visualise the level of misalignment, if any, in actual flight conditions, a system has been developed to measure the in-flight co-linearity of the generation and detection beams employed in the LURSOT system. The system employs a mirror to image the two laser beams onto a thin plate of blackened aluminum. The laser beams in turn, heat a spot on the aluminum plate, and an infrared camera (sensitive in the thermal infrared) captures an image of these ‘‘hot’’ spots. A suitable delay is added so that each beam can be observed sequentially. The image sequences of the IR camera are digitized and stored for processing in a
490
C.E. Brown, M.F. Fingas / Marine Pollution Bulletin 47 (2003) 485–492
computer. During flight, a mirror is moved into position to divert the laser beams onto the aluminum plate at any time (if required for verification of co-linearity). The positions of the two beams are subsequently determined
through data analysis of the infrared camera images and beam co-linearity/overlap are confirmed. The redesigned LURSOT system was assembled at IMI and extensively tested in a large-scale laboratory
Fig. 3. New LURSOT support structure illustrating telescope (lower left), CO2 laser (centre), beam co-linearity monitoring device (upper left) and focusing optics (right).
Fig. 4. New LURSOT support structure illustrating Doppler shift compensation device (upper left), detection laser (centre) and MISER controller (upper right).
C.E. Brown, M.F. Fingas / Marine Pollution Bulletin 47 (2003) 485–492
environment to confirm functionality of individual components and the successful measurement of actual oil slick thickness. Individual components of the LURSOT system in the new support structure are shown in Figs. 3 and 4.
4. Flight test program The LURSOT structure has now been transported to Environment CanadaÕs hangar facilities in Ottawa where it is being installed on ESTDÕs DC-3 aircraft and is currently undergoing a final certification process. Following installation, the LURSOT will undergo a series of flight tests. Following verification of acceptable airborne operation of individual LURSOT system components, a final set of test flights will focus on acquiring an airborne measurement of oil slick thickness. Initially, a signal will be collected over a large body of water, such as a lake, without an oil layer. The laser-ultrasonic signal would consist of the ‘‘surface signal’’ generated by the initial thermal expansion of the water surface, caused by the absorption of the generation laser. Following the successful collection of a laser-ultrasonic signal over a large body of water, flights over manmade pools with oil-on-water slicks will be undertaken. Laser-ultrasonic signals will be recorded and the thickness of the oil layers determined.
5. Recommended research and development The following is a list of recommendations for research and development needs to identify future advancements in the development and utilization of laserultrasonic technologies in remote sensing for detection and measuring thickness of spilled oil: • increase the sampling rate of the lasers employed to provide continuous flight-line coverage; • develop a longer wavelength detection laser to improve eye-safety for an operational system; and • decrease the overall system size for an operational system.
6. Summary and conclusions The LURSOT system provides a very accurate means of measuring oil thickness and has great potential as an airborne oil spill remote sensor. The technology is being developed by a consortium of agencies including Environment Canada, the Industrial Materials Institute, National Research Council of Canada, Imperial Oil Limited, and the United States Minerals Management
491
Service. Laboratory tests have confirmed the viability of the method and a test unit will be flown to confirm its operability on an airborne platform. LURSOT will provide an absolute measurement of oil thickness from an airborne platform. This information is necessary for the effective direction of spill countermeasures such as dispersant application and in situ burning. This information may also allow for the calibration of some of the more economical and readily available pieces of remote sensing equipment. Calibration of these wide fieldof-view sensors would provide a reliable method of estimating the volume of rogue oil slicks. Present airborne remote sensors often provide erroneous estimates of oil quantity. It is our hope that the information provided by this laser-based oil spill remote sensor will aid spill responders in their quest to mitigate the potentially disastrous effects of an oil spill on sensitive marine and coastal environments.
Acknowledgements The development of the LURSOT sensor has been supported by the following agencies: • Emergencies Science and Technology Division, Environment Canada, • Imperial Oil Resources Ltd., • United States Minerals Management Service, • Industrial Materials Institute, National Research Council of Canada, • United States Coast Guard and • American Petroleum Institute.
References Aussel, J.D., Monchalin, J.P., 1989. Laser-ultrasonic measurement of oil thickness on water from aircraft. Feasibility Study, Industrial Materials Research Institute Report. National Research Council of Canada, Boucherville, Que. Blouin, A., Monchalin, J.-P., 1994. Detection of ultrasonic motion of a scattering surface by two-wave mixing in a photorefractive GaAs crystal. Appl. Phys. Lett. 65 (8), 932–934. Brown, C.E., Fruhwirth, M., Fingas, M.F., Goodman, R.H., Choquet, M., Heon, R., Vaudreuil, G., Monchalin, J.P., Padioleau, C., 1994. Laser ultrasonic remote sensing of oil thickness: absolute measurement of oil slick thickness. In: Proceedings of the First International Airborne Remote Sensing Conference and Exhibition, vol. 1. Environmental Research Institute of Michigan, Ann Arbor, MI, pp. 567–578. Brown, C.E., Fingas, M.F., Goodman, R.H., Choquet, M., Blouin, A., Drolet, D., Monchalin, J.-P., Hardwick, C.D., 1997. The LURSOT sensor: providing absolute measurement of oil slick thickness. In: Proceedings of the Fourth Conference on Remote Sensing for Marine and Coastal Environments, vol. 1. Environmental Research Institute of Michigan, Ann Arbor, MI, USA, pp. I-393–I397.
492
C.E. Brown, M.F. Fingas / Marine Pollution Bulletin 47 (2003) 485–492
Brown, C.E., Fingas, M.F., 1999. Oil spill surveillance, monitoring and remote sensing: a global review. In: Proceedings of the Twentysecond Arctic and Marine Oil Spill Program Technical Seminar. Environment Canada, Ottawa, Ont, pp. 387–402. Brown, C.E., Fingas, M.F., Goodman, R.H., Mullin, J.V., Choquet, M., Monchalin, J.-P., 2000. Progress in achieving airborne oil slick thickness measurement. In: Proceedings of the Twenty-third Arctic and Marine Oil Spill Program Technical Seminar. Environment Canada, Ottawa, Ont, pp. 493–498. Brown, H.M., Goodman, R.H., 1986. In situ burning of oil in ice leads. In: Proceedings of the Ninth Annual Arctic and Marine Oilspill Program Technical Seminar. Environment Canada, Ottawa, Ont, pp. 245–256. Choquet, M., Monchalin, J.P., 1990. Modelling of thermoelastic generation of ultrasound in oil upon water. Industrial Materials Institute Internal Report. National Research Council of Canada, Boucherville, Que. Choquet, M., Heon, R., Vaudreuil, G., Monchalin, J.-P., Padioleau, C., Goodman, R.H., 1993. Remote thickness measurement of oil slicks on water by laser ultrasonics. In: Proceedings of the 1993 International Oil Spill Conference. American Petroleum Institute, Washington, DC, pp. 531–536. F€ ast, O., 1986. Remote sensing of oil on water–air and space-borne systems. In: Proceedings of the DOOS Seminar. SINTEF, Trondheim, Norway. Fingas, M.F., Brown, C.E., Mullin, J.V., 1998. The visibility limits of oil on water and remote sensing thickness detection limits. In: Proceedings of the Fifth Thematic Conference on Remote Sensing for Marine and Coastal Environments. Environmental Research Institute of Michigan (ERIM), Ann Arbor, MI, pp. II-411–II418. Goodman, R.H., Fingas, M.F., 1988. The use of remote sensing for the determination of dispersant effectiveness. In: Proceedings of the Eleventh Arctic and Marine Oilspill Program Technical Seminar. Environment Canada, Ottawa, Ont, pp. 377–384. Goodman, R., 1994. Remote sensing resolution and oil slick inhomogeneities. In: Proceedings of the Second Thematic Conference: Remote Sensing for Marine and Coastal Environments, vol. 1. Environmental Research Institute of Michigan, Ann Arbor, MI, pp. 1–6. Ing, R.K., Monchalin, J.-P., 1991. Broadband optical detection of ultrasound by two-wave mixing in a photorefractive crystal. Appl. Phys. Lett. 59 (25), 3233–3235. Krapez, J.C., Cielo, P., 1992. Optothermal evaluation of oil film thickness. J. Appl. Phys. 72 (4), 1255–1261. McMahon, O.B., Murphy, T.J., Brown, E.R., 1997. Remote measurement of oil spill thickness. In: Proceedings of the Fourth Thematic Conference on Remote Sensing for Marine and Coastal Environ-
ments, vol. I. Environmental Research Institute of Michigan (ERIM), Ann Arbor, MI, pp. 353–360. Monchalin, J.P., 1986. Optical detection of ultrasound. IEEE Trans. Ultrason. Ferroelectr. Frequency Control UFFC-33 (5), 485–499. Mussetto, M.S., Yujiri, L., Dixon, D.P., Hauss, B.I., Eberhard, C.D., 1994. Passive millimeter wave radiometric sensing of oil spills. In: Proceedings of the Second Thematic Conference on Remote Sensing for Marine and Coastal Environments: Needs, Solutions and Applications. Environmental Research Institute of Michigan (ERIM), Ann Arbor, MI, pp. I-35–I-46. OÕNeil, R.A., Neville, R.A., Thompson, V., 1983. The Arctic Marine Oilspill Program (AMOP) remote sensing study. Report EPS 4-EC83-3. Environment Canada, Ottawa, ON, 257 p. Patsayeva, S., Yuhakov, V., Varlamov, V., Barbini, R., Fantoni, R., Frassanito, C., Palucci, A., 2000. Laser spectroscopy of mineral oils on the water surface. In: Proceedings of the Fourth EARSeL Workshop Lidar Remote Sensing of Land and Sea, Dresden, Germany. Available from, 9 p. Pelyushenko, S.A., 1995. Microwave radiometer system for the detection of oil slicks. Spill Sci. Technol. Bull. 2 (4), 249–254. Pelyushenko, S.A., 1997. The use of microwave radiometer scanning system for detecting and identification of oil spills. In: Proceedings of the Fourth Thematic Conference on Remote Sensing for Marine and Coastal Environments, vol. I. Environmental Research Institute of Michigan (ERIM), Ann Arbor, MI, pp. 381–385. Reimer, E.R., Rossiter, J.R., 1987. Measurement of oil thickness on water from aircraft. A: Active microwave spectroscopy; B: Electromagnetic thermoelastic emission, Environmental Studies Revolving Fund Report Number 078, Ottawa, Ont. Skou, N., Sorensen, B.M., Poulson, A., 1994. A new airborne dual frequency microwave radiometer for mapping and quantifying mineral oil on the sea surface. In: Proceedings of the Second Thematic Conference on Remote Sensing for Marine and Coastal Environments. Environmental Research Institute of Michigan (ERIM), Ann Arbor, MI, pp. II-559–II-565. S€ uss, H., Gr€ uner, K., Wilson, W.J., 1989. Passive millimeter wave imaging: a tool for remote sensing. Alta Frequenza LVIII (5–6), 457–465. Ulaby, F.T., Moore, R.K., Fung, A.K., 1989. Microwave Remote Sensing: Active and Passive. Artech House Inc, Norwood, MA. pp. 1466–1479. Wang, Z., Nur, A., 1991. Ultrasonic velocity in pure hydrocarbons and mixtures. J. Acoust. Soc. Am. 89, 2725–2730. Zhifu, S., Wiesbeck, W., 1988. A study of passive microwave remote sensing. In: Proceedings of the 1988 International Geoscience and Remote Sensing Symposium. European Space Agency, Paris, France, pp. 1091–1094.
Technical Note
Development of airborne oil thickness measurements Carl E. Brown *, Mervin F. Fingas
1
Emergencies Science and Technology Division, Environment Canada, 335 River Road, Ottawa, Ont., Canada, K1A 0H3
Abstract A laboratory sensor has now been developed to measure the absolute thickness of oil on water slicks. This prototype oil slick thickness measurement system is known as the laser-ultrasonic remote sensing of oil thickness (LURSOT) sensor. This laser optoacoustic sensor is the initial step in the ultimate goal of providing an airborne sensor with the ability to remotely measure oilon-water slick thickness. The LURSOT sensor employs three lasers to produce and measure the time-of-flight of ultrasonic waves in oil and hence provide a direct measurement of oil slick thickness. The successful application of this technology to the measurement of oil slick thickness will benefit the scientific community as a whole by providing information about the dynamics of oil slick spreading and the spill responder by providing a measurement of the effectiveness of spill countermeasures such as dispersant application and in situ burning. This paper will provide a review of early developments and discuss the current state-of-the-art in the field of oil slick thickness measurement. 2003 Published by Elsevier Ltd. Keywords: Oil; Slick; Thickness; Measurement; Laser; Opto-acoustic
1. Introduction Scientists and spill response personnel have long been searching for an accurate way to measure oil-on-water slick thickness. Until recently there was no reliable laboratory or field method for the precise measurement of oil-on-water slick thickness. Knowledge of slick thickness will result in more effective direction of oil spill countermeasures including dispersant application and in situ burning. In reality, the effectiveness of specific dispersants could be determined quantitatively by the accurate measurement of the oil remaining on the water surface following dispersant application (Goodman and Fingas, 1988). Furthermore, the ability to measure oil slick thickness should provide significant advances to the fundamental understanding of the dynamics of oil slick spreading. Finally, there is a need to calibrate some of the more economical and readily available pieces of remote sensing equipment. Several of these sensors provide relative indications of slick thickness, i.e., whe* Corresponding author. Tel.: +1-613-991-1118; fax: +1-613-9919485. E-mail addresses: [email protected] (C.E. Brown), merv.fi[email protected] (M.F. Fingas). 1 Tel.: +1-613-998-9622.
0025-326X/$ - see front matter 2003 Published by Elsevier Ltd. doi:10.1016/S0025-326X(03)00203-0
ther the slick is thick or thin. Calibration of these wide field-of-view sensors would provide a reliable method of estimating the volume of rogue oil slicks. Present airborne visual surveillance of oil slicks often produces erroneous estimates of oil quantity.
2. Early developments Several remote sensing techniques have been evaluated over the past 15 years for their ability to measure oil-on-water slick thickness. Laser fluorosensors are well known for their ability to positively detect and classify petroleum oils (Brown and Fingas, 1999). In clear waters, the ultraviolet laser beams employed in airborne laser fluorosensor systems stimulate Raman scattering in addition to fluorescence (when oil is present). With excitation at 308 nm (XeCl laser), the water Raman scattering peak is observed at 344 nm. When petroleum oils of sufficient thickness (typically >10–20 lm) are present on the surface of the water, the ultraviolet laser energy is completely absorbed by the oil and does not penetrate into the water column, hence no Raman scattering signal is observed. At oil thicknesses less than 10–20 lm, some of the ultraviolet light penetrates the water and
486
C.E. Brown, M.F. Fingas / Marine Pollution Bulletin 47 (2003) 485–492
induces the Raman scattering. With knowledge of certain properties of the oil (including the wavelength specific oil absorption coefficient) one can estimate the thickness of the oil slick from the degree of suppression of the Raman signal. While there has been some additional work with laser fluorosensors (Patsayeva et al., 2000), the suppression of the water Raman peak has not been fully exploited or tested. The technique works for thin slicks, but not necessarily for thick ones, at least not with a single excitation wavelength. Oil, which is optically thick, absorbs solar radiation and re-emits a portion of this radiation as thermal energy, primarily in the infrared region of the electromagnetic spectrum at 8–14 lm. In infrared (IR) images, thick oil appears hot, intermediate thicknesses of oil appear cool, and thin oil or sheens are not detected. The thicknesses at which these transitions occur are not known, but evidence indicates that the transition between the hot and cold layer lies between 50 and 150 lm and the minimum detectable layer is between 20 and 70 lm (Fingas et al., 1998). The reason for the appearance of the ‘‘cool’’ slick is not fully understood. The most plausible theory is that a moderately thin layer of oil on the water surface causes destructive interference of the thermal radiation waves emitted by the water, thereby reducing the amount of thermal radiation emitted by the water. Attempts have been made to calibrate the thickness appearance of infrared imagery, but without success. It is suspected that the temperatures of the slick as seen in the infrared are highly dependent on oil type, sun angle, and weather conditions. If so, it may not be possible to use the infrared as a calibrated tool for measuring thickness. The ocean reflects interstellar microwave radiation. Oil on the ocean is a stronger reflector of microwave radiation than the water and thus appears as a bright object on a darker sea. The reflectivity (or apparent emissivity) factor of water is 0.4 compared to 0.8 for oil (OÕNeil et al., 1983; Ulaby et al., 1989). A passive device can detect this difference in emissivity and could therefore be used to detect oil. In addition, as the signal changes with thickness, in theory, the device could be used to measure thickness. This detection method has not been very successful in the field, however, as several environmental and oil-specific parameters must be known a priori. Furthermore, the signal return is dependent on oil thickness but in a cyclical fashion. A given signal strength can imply any one of two or three signal film thicknesses within a given slick. Microwave energy emission is greatest when the effective thickness of the oil equals an odd multiple of one quarter of the wavelength of the observed energy. Biogenic materials also interfere and the signal-to-noise ratio is low. It is also difficult to achieve high spatial resolution with microwave radiometers (Goodman, 1994). The Swedish Space Agency has conducted studies with a variety of
systems, including a dual-band, 22.4- and 31-GHz device, and a single-band 37-GHz device (F€ast, 1986). Skou et al. (1994) describe a two-channel device operating at 37.5 and 10.7 GHz. Mussetto et al. (1994) at TRW have tested 44-94-GHz and 94-154-GHz, twochannel devices over oil slicks. This work showed that the correlation with slick thickness is poor and suggested that factors other than thickness also change surface brightness. They also suggested that a single-channel device might be useful as an all-weather, relative-thickness instrument. Tests of single-channel devices over oil slicks have also been described in the literature, specifically a 36-GHz (Zhifu and Wiesbeck, 1988) and a 90-GHz device (S€ uss et al., 1989). A new method of microwave radiometry has recently been developed in which the polarization contrasts at two orthogonal polarizations are measured in an attempt to measure oil slick thickness (Pelyushenko, 1995, 1997). A series of frequency-scanning radiometers has been built and appears to have overcome the difficulties with the cyclical behaviour (McMahon et al., 1997). In summary, passive microwave radiometers may have potential as all-weather oil sensors. Their potential as a reliable device for measuring slick thickness, however, is uncertain at this time. The signal strength measured by microwave radiometers can imply one of several thicknesses. Microwave radiometers therefore, do not appear to have potential, other than for measuring relative oil thickness. Several optical, electrical and acoustic techniques for measuring oil thickness have been studied (Reimer and Rossiter, 1987). Two promising techniques were investigated in a series of laboratory measurements. In the first technique, known as ‘‘thermal mapping’’, a laser is used to heat a region of oil and the resultant temperature profiles created over a small region near this heating are examined using an infrared camera (Krapez and Cielo, 1992). The temperature profiles created are dependent on the oil thickness. A more promising technique involving laser acoustics is described in the following section (Aussel and Monchalin, 1989; Choquet et al., 1993). As accurate surface methods do not exist, it is therefore very difficult to calibrate existing equipment (Brown and Goodman, 1986). The use of sorbent techniques to measure surface thickness yields highly variable results (Goodman and Fingas, 1988).
3. Laser-ultrasonic remote sensing of oil thickness The laser-ultrasonic remote sensing of oil thickness (LURSOT) sensor is a three-laser system with one of the lasers coupled to an optical interferometer to accurately measure oil thickness (Brown et al., 1994, 2000; Choquet et al., 1993). The measurement process is initiated with a
C.E. Brown, M.F. Fingas / Marine Pollution Bulletin 47 (2003) 485–492
thermal pulse created in the oil layer by the absorption of a powerful infrared carbon dioxide (CO2 ) laser pulse. Rapid thermal expansion of the oil occurs near the surface where the CO2 laser beam was absorbed. This causes a step-like rise of the sample surface and creates a high frequency and large bandwidth (15 MHz for oil) acoustic pulse. The acoustic pulse moves down through the oil until it reaches the oil–water interface, where it is partially transmitted (86%) and partially reflected back towards the oil–air interface (14%), where it produces a slight displacement of the oil surface. The time required for the acoustic pulse to travel through the oil and back to the surface is a function of the oil thickness and the acoustic velocity of the oil. The displacement of the surface is measured by a second laser probe beam aimed at the surface. The motion of the surface causes a phase or frequency shift (Doppler shift) in the reflected probe beam. This modulation of the probe beam is then demodulated with an interferometer (Monchalin, 1986). The interferometer behaves like a narrow optical filter which directly demodulates the light when the laser probe frequency is set on one of the slopes of the filter response (half height). The technique is sensitive to the high frequency surface displacement caused by the ultrasonic pulse, and is insensitive to low frequency motions such as vibration. The LURSOT system uses a third laser (a continuous wave HeNe laser) to interrogate the water surface and generate a trigger pulse when the correct surface geometry for thickness measurement exists. The strong absorption of the infrared laser radiation by oil, combined with the high thermal coefficient of expansion of oil results in the efficient generation of ultrasonic pulses. This efficient generation of ultrasound should theoretically allow operation of the sensor from an airborne platform since the laser beam does not have to be tightly focussed to produce a sufficient energy density. Problems arise however, with the weak acoustic impedance mismatch between oil and water which results in a weak acoustic reflection coefficient. Calculations indicate a reflection coefficient of only 14%, based on typical acoustic parameters of oil and water (see Table 1). The low reflection coefficient leads to low amplitude acoustic pulse echoes. Therefore, high detection sensitivity is needed to observe the small surface deflections. The detection sensitivity is directly related to the amount of light received by the detector (Monchalin, 1986). As the probe beam illuminates an essentially flat surface, the reflected beam has low divergence and can be collected with a simple optical telescope of proper aperture. A very high-intensity probe laser pulse is employed to compensate for the loss of probe beam intensity as a result of the low reflection coefficient of the oil at 1.06 lm. The accuracy of the laser-ultrasonic measurement of oil thickness is dependent on both the precision of the
487
Table 1 Acoustic and optical properties of Norman Wells crude oil and water Property
Norman Wells crude
Water
Acoustic impedance (Pa s 1 ) Acoustic velocity (m s 1 ) Density (g ml 1 ) Specific heat (J kg 1 K 1 ) Thermal expansion coefficient (K 1 ) Optical penetration depth at 10.6 lm Optical reflectivity at 1.06 lm
1:13 106 1410 0.8 1000 1:0 10 4 100 0.045
1:5 106 1500 1 4128 4:1 10 5 10 0.02
measurement of the time delay between surface displacement echoes and the accuracy of the measurement of the acoustic velocity of the oil. The acoustic velocity of oil can be measured very accurately in the laboratory using conventional acoustic techniques. Variations in the chemical makeup of various petroleum products will lead to small differences in the acoustic velocities of these products (Wang and Nur, 1991). There is a slight dependence of the acoustic velocity on temperature; however, investigations at the Industrial Materials Institute (IMI) of the National Research Council of Canada indicate the variation is <3% for a temperature change from 5 to 15 C. In an emergency response to a marine oil spill, the type and degree of weathering of the oil may not be known. It is therefore the uncertainty of the acoustic velocity of the oil that limits the accuracy of the remote thickness measurement. This expected uncertainty is, however, more than acceptable for making effective oilspill countermeasure decisions (Goodman, 1994). The LURSOT system has made a series of laboratory measurements on a range of thicknesses of weathered Norman Wells crude oil on water. Oil thicknesses from 250 lm to 35 mm have been measured over distances ranging from 2 to 91 m. In these experiments, care was taken to ensure that sufficient water (at least 4 cm) was placed in the container to prevent any interference echo from the water–container interface. A pulsed CO2 laser (Laser Science, 200 mJ, 100 ns pulse width) was used for generation of the ultrasound pulses in the oil. The probe laser employed was a frequency-stabilized Nd:YAG (1.06 lm, 50 ls pulse width) developed by IMI and UltraOptec Incorporated. In order to provide maximum sensitivity, the peak amplitude of the probe laser beam was adjusted to nearly saturate the detector. A single lens was used to focus the probe laser beam to a spot approximately 4 mm in diameter, equal to that of the unfocussed CO2 pump laser beam. The pump and probe beams were combined with a germanium dichroic optic. Over short distances (a couple of metres), a 15-cm diameter lens was used to collect the reflected probe beam and focus the light into an optical fiber, which transmits the light to a confocal Fabry–Perot Interferometer (c-FPI). A typical laser-ultrasonic signal (average of 30 consecutive pulses) recorded in these experiments is
488
C.E. Brown, M.F. Fingas / Marine Pollution Bulletin 47 (2003) 485–492
Fig. 1. Laser-ultrasonic signal, average of 30 data curves, of an oil layer on water-direct thickness observation: 5 mm (50%), laser-ultrasonic measurement: 6.23 mm (1%).
shown in Fig. 1. Note that the surface displacement signal just barely saturates the detector. In addition to the surface signal, the first and second echoes are clearly observable. The shapes of the surface displacement and the echoes are different, as expected. The second echo is inverted with respect to the first because of the higher acoustic impedance of water compared to oil, giving an inverted reflection (for displacement). The amplitude of the reflected acoustic pulse is about 12%, in good agreement with the predicted 14% based on properties listed in Table 1. Owing to the consistency of the shapes of the echoes, a direct cross-correlation of the estimate of the time delay is possible. Using the measured acoustic velocity of 1410 m s 1 for Norman Wells crude, the thickness of oil measured by laser-ultrasonics and cross-correlation of the first and second echoes is 6.23 mm (1%). The 1% accuracy is in reference to the measurement of the acoustic velocity of Norman Wells crude oil using conventional ultrasonics. Physical measurement of the thickness gives a value of 5 mm (50%), with the 50% error a result of capillary wicking of the oil on the wall of the beaker. In actual remote sensing operations, the signal-to-noise ratio may be insufficient to allow observation of the second echo. Cross-correlation of the surface signal and the first echo is possible, with a slight error introduced because of the different shapes of the signals. The thickness of the oil layer is determined as 6.26 mm, using the surface signal and the first echo, a difference of 0.5% from that determined using the first and second echoes. In general, the difference associated with using the surface signal
and the first echo is less than 1% when compared to use of two consecutive echoes (Choquet and Monchalin, 1990). Several modifications to the LURSOT system have been made in preparation for a second airborne test of the complete system. An optimized flashlamp system was developed to reduce the turn-on time of the probe laser to approximately 120 ls. This extended the amount of time available for thickness measurement to 80 ls, permitting measurement of oil slick thicknesses up to 6 cm (assuming an acoustic velocity of 1500 m s 1 ). To supply the required power for the Nd:YAG probe laser, an uninterruptible power source will be employed. Researchers at IMI have investigated alternatives to the c-FPI that would maintain high sensitivity while reducing the susceptibility of the system to the vibrations and acoustic noise associated with operation of the c-FPI in an airborne environment. Recently, a new method for the detection of ultrasound by two-wave mixing in a photorefractive crystal was developed at IMI (Ing and Monchalin, 1991; Blouin and Monchalin, 1994). This method is an adaptive as opposed to a passive interferometric technique. Broad detection bandwidth is achieved through interference of the wave scattered from the surface in motion with a sidebandfree reference wave. This reference wave should have a wave front closely matching that of the wave scattered from the surface. As illustrated in Fig. 2, the beam scattered from the surface (signal) and a beam obtained directly from the laser (pump) interfere inside a photorefractive crystal (Blouin and Monchalin, 1994). This
C.E. Brown, M.F. Fingas / Marine Pollution Bulletin 47 (2003) 485–492
Fig. 2. Basic experimental setup for optical detection of ultrasound by two-wave mixing in a GaAs crystal. P is a polarizer and PBS is a polarizing beam splitter, both being oriented at 45 to the plane of incidence, BS is a beam splitter.
interference temporarily creates an index of refraction grating as a result of the photorefractive effect. This grating diffracts the pump beam into a reference beam with a beam closely matching the wave front of the signal (scattered) wave. The resulting reference beam is sideband-free, since it is derived directly from the laser and since the grating does not follow the rapid ultrasonic phase perturbations. The method provides a reference beam with sufficient intensity to give quantum limited detection by producing efficient coupling between the pump and signal beams. This can be achieved when the index of refraction grating is p=2 phase shifted relative to the interference grating. This results in the signal wave and the diffracted pump wave (reference) being in phase at the exit of the crystal. This phase relationship leads to weak and quadratic detection of small phase modulations, such as those produced by ultrasonic motion. To produce linear and sensitive detection, two polarizations are given to the signal wave and a retardation plate (Babinet–Soleil compensator) is added at the output, so that the reference and transmitted waves are in quadrature (Blouin and Monchalin, 1994). The resulting two signals are of opposite sign, which when combined in a differential amplifier, leads to a doubling of the signal. Recently, the measurement of the thickness of a layer of Norman Wells crude oil on water using an ultrasonic detection system based on two-wave mixing in a photorefractive GaAs crystal was demonstrated at IMI. The new detector has been called the photorefractive optical ultrasonic detector (PROUD). Following the initial demonstration of the oil slick thickness measurement, the PROUD system was shaken vigorously using a mechanical shaker at several frequencies and amplitudes. The ability of the PROUD system to measure oil thickness was not at all affected by this motion. The shaking motion was far more severe than one would ever expect in an airborne environment, leading one to believe that PROUD will be more suitable for use in an airborne LURSOT system than the c-FPI. The PROUD detector is sensitive to frequency changes caused by
489
motion of the target relative to the laser source (in the aircraft). To compensate for these frequency (Doppler) shifts, an optical frequency compensation device (OFCD) was devised and tested. In order to employ the OFCD on board an aircraft, a measurement of vertical velocity must be provided. This was accomplished through the development of a vertical velocity sensor (VVS) by the Institute for Aerospace Research at the National Research Council of Canada, in Ottawa. The VVS couples a differential global positioning system receiver with an accelerometer in order to provide accurate real-time vertical velocity measurement. The VVS was tested onboard the DC-3 and found to operate satisfactorily when isolated mechanically from the floor of the aircraft and filtered appropriately (Brown et al., 1997). A decision was made to construct a device to measure the instantaneous optical frequency of the returning target laser beam. This device provides a diagnostic as to how well the OFCD is working. Also, this device will provide data on the severity of the Doppler shift induced on the probe laser during normal test conditions of the LURSOT. The broad range of temperatures encountered in the aircraft was found to induce differences within the optical beam path of the probe laser (Nd:YAG) that were outside acceptable limits. To correct this, a novel compact laser system was developed and mounted on a zero thermal expansion carbon-epoxy optical breadboard. In addition to the known problems associated with operation in the airborne environment, there were concerns about the possible loss of co-linearity of the laser beams employed in the LURSOT system. A complete theoretical analysis of the support structure which houses the optical and laser components was undertaken at the University of TorontoÕs Aerospace Institute in order to understand the effect of vibrations on the colinearity of the optical system. The investigation uncovered several shortcomings with the initial structure and provided the information required to design a structure that provides the stability required to ensure co-linearity of the laser beams at the operating altitude of 300 ft. The structure was subsequently constructed under contract by Aerotech Incorporated, Laval, Quebec. In order to better visualise the level of misalignment, if any, in actual flight conditions, a system has been developed to measure the in-flight co-linearity of the generation and detection beams employed in the LURSOT system. The system employs a mirror to image the two laser beams onto a thin plate of blackened aluminum. The laser beams in turn, heat a spot on the aluminum plate, and an infrared camera (sensitive in the thermal infrared) captures an image of these ‘‘hot’’ spots. A suitable delay is added so that each beam can be observed sequentially. The image sequences of the IR camera are digitized and stored for processing in a
490
C.E. Brown, M.F. Fingas / Marine Pollution Bulletin 47 (2003) 485–492
computer. During flight, a mirror is moved into position to divert the laser beams onto the aluminum plate at any time (if required for verification of co-linearity). The positions of the two beams are subsequently determined
through data analysis of the infrared camera images and beam co-linearity/overlap are confirmed. The redesigned LURSOT system was assembled at IMI and extensively tested in a large-scale laboratory
Fig. 3. New LURSOT support structure illustrating telescope (lower left), CO2 laser (centre), beam co-linearity monitoring device (upper left) and focusing optics (right).
Fig. 4. New LURSOT support structure illustrating Doppler shift compensation device (upper left), detection laser (centre) and MISER controller (upper right).
C.E. Brown, M.F. Fingas / Marine Pollution Bulletin 47 (2003) 485–492
environment to confirm functionality of individual components and the successful measurement of actual oil slick thickness. Individual components of the LURSOT system in the new support structure are shown in Figs. 3 and 4.
4. Flight test program The LURSOT structure has now been transported to Environment CanadaÕs hangar facilities in Ottawa where it is being installed on ESTDÕs DC-3 aircraft and is currently undergoing a final certification process. Following installation, the LURSOT will undergo a series of flight tests. Following verification of acceptable airborne operation of individual LURSOT system components, a final set of test flights will focus on acquiring an airborne measurement of oil slick thickness. Initially, a signal will be collected over a large body of water, such as a lake, without an oil layer. The laser-ultrasonic signal would consist of the ‘‘surface signal’’ generated by the initial thermal expansion of the water surface, caused by the absorption of the generation laser. Following the successful collection of a laser-ultrasonic signal over a large body of water, flights over manmade pools with oil-on-water slicks will be undertaken. Laser-ultrasonic signals will be recorded and the thickness of the oil layers determined.
5. Recommended research and development The following is a list of recommendations for research and development needs to identify future advancements in the development and utilization of laserultrasonic technologies in remote sensing for detection and measuring thickness of spilled oil: • increase the sampling rate of the lasers employed to provide continuous flight-line coverage; • develop a longer wavelength detection laser to improve eye-safety for an operational system; and • decrease the overall system size for an operational system.
6. Summary and conclusions The LURSOT system provides a very accurate means of measuring oil thickness and has great potential as an airborne oil spill remote sensor. The technology is being developed by a consortium of agencies including Environment Canada, the Industrial Materials Institute, National Research Council of Canada, Imperial Oil Limited, and the United States Minerals Management
491
Service. Laboratory tests have confirmed the viability of the method and a test unit will be flown to confirm its operability on an airborne platform. LURSOT will provide an absolute measurement of oil thickness from an airborne platform. This information is necessary for the effective direction of spill countermeasures such as dispersant application and in situ burning. This information may also allow for the calibration of some of the more economical and readily available pieces of remote sensing equipment. Calibration of these wide fieldof-view sensors would provide a reliable method of estimating the volume of rogue oil slicks. Present airborne remote sensors often provide erroneous estimates of oil quantity. It is our hope that the information provided by this laser-based oil spill remote sensor will aid spill responders in their quest to mitigate the potentially disastrous effects of an oil spill on sensitive marine and coastal environments.
Acknowledgements The development of the LURSOT sensor has been supported by the following agencies: • Emergencies Science and Technology Division, Environment Canada, • Imperial Oil Resources Ltd., • United States Minerals Management Service, • Industrial Materials Institute, National Research Council of Canada, • United States Coast Guard and • American Petroleum Institute.
References Aussel, J.D., Monchalin, J.P., 1989. Laser-ultrasonic measurement of oil thickness on water from aircraft. Feasibility Study, Industrial Materials Research Institute Report. National Research Council of Canada, Boucherville, Que. Blouin, A., Monchalin, J.-P., 1994. Detection of ultrasonic motion of a scattering surface by two-wave mixing in a photorefractive GaAs crystal. Appl. Phys. Lett. 65 (8), 932–934. Brown, C.E., Fruhwirth, M., Fingas, M.F., Goodman, R.H., Choquet, M., Heon, R., Vaudreuil, G., Monchalin, J.P., Padioleau, C., 1994. Laser ultrasonic remote sensing of oil thickness: absolute measurement of oil slick thickness. In: Proceedings of the First International Airborne Remote Sensing Conference and Exhibition, vol. 1. Environmental Research Institute of Michigan, Ann Arbor, MI, pp. 567–578. Brown, C.E., Fingas, M.F., Goodman, R.H., Choquet, M., Blouin, A., Drolet, D., Monchalin, J.-P., Hardwick, C.D., 1997. The LURSOT sensor: providing absolute measurement of oil slick thickness. In: Proceedings of the Fourth Conference on Remote Sensing for Marine and Coastal Environments, vol. 1. Environmental Research Institute of Michigan, Ann Arbor, MI, USA, pp. I-393–I397.
492
C.E. Brown, M.F. Fingas / Marine Pollution Bulletin 47 (2003) 485–492
Brown, C.E., Fingas, M.F., 1999. Oil spill surveillance, monitoring and remote sensing: a global review. In: Proceedings of the Twentysecond Arctic and Marine Oil Spill Program Technical Seminar. Environment Canada, Ottawa, Ont, pp. 387–402. Brown, C.E., Fingas, M.F., Goodman, R.H., Mullin, J.V., Choquet, M., Monchalin, J.-P., 2000. Progress in achieving airborne oil slick thickness measurement. In: Proceedings of the Twenty-third Arctic and Marine Oil Spill Program Technical Seminar. Environment Canada, Ottawa, Ont, pp. 493–498. Brown, H.M., Goodman, R.H., 1986. In situ burning of oil in ice leads. In: Proceedings of the Ninth Annual Arctic and Marine Oilspill Program Technical Seminar. Environment Canada, Ottawa, Ont, pp. 245–256. Choquet, M., Monchalin, J.P., 1990. Modelling of thermoelastic generation of ultrasound in oil upon water. Industrial Materials Institute Internal Report. National Research Council of Canada, Boucherville, Que. Choquet, M., Heon, R., Vaudreuil, G., Monchalin, J.-P., Padioleau, C., Goodman, R.H., 1993. Remote thickness measurement of oil slicks on water by laser ultrasonics. In: Proceedings of the 1993 International Oil Spill Conference. American Petroleum Institute, Washington, DC, pp. 531–536. F€ ast, O., 1986. Remote sensing of oil on water–air and space-borne systems. In: Proceedings of the DOOS Seminar. SINTEF, Trondheim, Norway. Fingas, M.F., Brown, C.E., Mullin, J.V., 1998. The visibility limits of oil on water and remote sensing thickness detection limits. In: Proceedings of the Fifth Thematic Conference on Remote Sensing for Marine and Coastal Environments. Environmental Research Institute of Michigan (ERIM), Ann Arbor, MI, pp. II-411–II418. Goodman, R.H., Fingas, M.F., 1988. The use of remote sensing for the determination of dispersant effectiveness. In: Proceedings of the Eleventh Arctic and Marine Oilspill Program Technical Seminar. Environment Canada, Ottawa, Ont, pp. 377–384. Goodman, R., 1994. Remote sensing resolution and oil slick inhomogeneities. In: Proceedings of the Second Thematic Conference: Remote Sensing for Marine and Coastal Environments, vol. 1. Environmental Research Institute of Michigan, Ann Arbor, MI, pp. 1–6. Ing, R.K., Monchalin, J.-P., 1991. Broadband optical detection of ultrasound by two-wave mixing in a photorefractive crystal. Appl. Phys. Lett. 59 (25), 3233–3235. Krapez, J.C., Cielo, P., 1992. Optothermal evaluation of oil film thickness. J. Appl. Phys. 72 (4), 1255–1261. McMahon, O.B., Murphy, T.J., Brown, E.R., 1997. Remote measurement of oil spill thickness. In: Proceedings of the Fourth Thematic Conference on Remote Sensing for Marine and Coastal Environ-
ments, vol. I. Environmental Research Institute of Michigan (ERIM), Ann Arbor, MI, pp. 353–360. Monchalin, J.P., 1986. Optical detection of ultrasound. IEEE Trans. Ultrason. Ferroelectr. Frequency Control UFFC-33 (5), 485–499. Mussetto, M.S., Yujiri, L., Dixon, D.P., Hauss, B.I., Eberhard, C.D., 1994. Passive millimeter wave radiometric sensing of oil spills. In: Proceedings of the Second Thematic Conference on Remote Sensing for Marine and Coastal Environments: Needs, Solutions and Applications. Environmental Research Institute of Michigan (ERIM), Ann Arbor, MI, pp. I-35–I-46. OÕNeil, R.A., Neville, R.A., Thompson, V., 1983. The Arctic Marine Oilspill Program (AMOP) remote sensing study. Report EPS 4-EC83-3. Environment Canada, Ottawa, ON, 257 p. Patsayeva, S., Yuhakov, V., Varlamov, V., Barbini, R., Fantoni, R., Frassanito, C., Palucci, A., 2000. Laser spectroscopy of mineral oils on the water surface. In: Proceedings of the Fourth EARSeL Workshop Lidar Remote Sensing of Land and Sea, Dresden, Germany. Available from