Field Experiments of Multi-Channel Oceanographic Fluorescence Lidar for Oil Spill and Chlorophyll-aDetection

LI Xiaolong, ZHAO Chaofang, MA Youjun and LIU Zhishen

1) Ocean Remote Sensing Institute, Ocean University of China, Qingdao 266003, P. R. China

2) Institute of Oceanology, Chinese Academy of Sciences, Qingdao 266071, P. R. China

Field Experiments of Multi-Channel Oceanographic Fluorescence Lidar for Oil Spill and Chlorophyll-aDetection

LI Xiaolong1),2), ZHAO Chaofang1),*, MA Youjun1), and LIU Zhishen1)

1) Ocean Remote Sensing Institute, Ocean University of China, Qingdao 266003, P. R. China

2) Institute of Oceanology, Chinese Academy of Sciences, Qingdao 266071, P. R. China

A Multi-channel Oceanographic Fluorescence Lidar (MOFL), with a UV excitation at 355 nm and multiple receiving channels at typical wavelengths of fluorescence from oil spills and chlorophyll-a(Chl-a), has been developed using the Laser- induced Fluorescence (LIF) technique. The sketch of the MOFL system equipped with a compact multi-channel photomultiplier tube (MPMT) is introduced in the paper. The methods of differentiating the oil fluorescence from the background water fluorescence and evaluating the Chl-aconcentration are described. Two field experiments were carried out to investigate the field performance of the system,i.e., an experiment in coastal areas for oil pollution detection and an experiment over the Yellow Sea for Chl-amonitoring. In the coastal experiment, several oil samples and other fluorescence substances were used to analyze the fluorescence spectral characteristics for oil identification, and to estimate the thickness of oil films at the water surface. The experiment shows that both the spectral shape of fluorescence induced from surface water and the intensity ratio of two channels (I495/I405) are essential to determine oilspill occurrence. In the airborne experiment, MOFL was applied to measure relative Chl-aconcentrations in the upper layer of the ocean. A comparison of relative Chl-aconcentration measurements by MOFL and the Moderate Resolution Imaging Spectroradiometer (MODIS) indicates that the two datasets are in good agreement. The results show that the MOFL system is capable of monitoring oil spills and Chl-ain the upper layer of ocean water.

oceanographic lidar; oil spill; marine environment; fluorescence spectrum; Raman scattering

1 Introduction

Monitoring marine environment is always of great concern for scientific community. In the upper layer of the ocean, oil slicks and chlorophyll are two key fluorescence substances related to ocean environment (Babichenko, 2008). The surveillance of oil pollution contributes significantly to marine ecological protection. Compared with passive sensors,i.e., UV, visible and IR sensors, multi

channel oceanographic Lidar (Light detection and ranging) takes advantage of the efficient, real-time, and all-day operation to monitor oil spill, and supports detection and identification of oil slicks by utilizing Laser Induced Fluorescence (LIF) (Jhaet al., 2008; Utkinet al., 2011). In the early 1970s, the first airborne laser fluorosensor was flown to map the extent of oil slicks (Brown and Fingas, 2003a). Another oil spill remote sensing program to measure oil slicks and analyze fluorescence properties of oil was reported in 1974 (Hengsterman and Reuter, 1990). Currently, most Lidar systems applied to detect oil spills employ a laser operated in the ultraviolet spectrum (300–355 nm), such as NASA’s Airborne Oceanographic Lidar (AOL), Environment Canada’s Scanning Laser Environmental Airborne Fluorosensor (SLEAF), the airborne laser fluorosensor (LFS) designed by the University of Oldenburg, Germany (Brown and Fingas, 2003), and Fluorescent Laser Spectrometer (FLS) developed by LDI Laser Diagnostic Instruments, Estonia (Yarovenkoet al., 2011).

As the key parameter of biomass and primary productivity, the concentration of Chl-ain the surface water can be measured relatively accurately by satellite ocean color sensors over open waters, but cannot be accurately obtained in coastal zones because of light absorption by suspended sediments (Liuet al., 2008). In contrast, the fluorescence Lidar can accurately detect Chl-afluorescence at 685 nm in coastal areas and becomes the most commonly used system for ocean color measurements (Hogeet al., 2005; Vasilescuet al., 2009).

In this paper, a Multi-channel Oceanographic Fluorescence Lidar (MOFL) system with an excitation laser at 355 nm and a compact 24-channel photomultiplier tube (MPMT) is designed to measure the fluorescence of oil pollution and chlorophyll. Section 2 describes the MOFL system and its components. Section 3 describes the methods used to detect oil slicks and estimate film thick-ness and Chl-aconcentration. Section 4 presents and discusses the results of the field experiments. In the conclusion, the feasibility of MOFL for oil spill detection and Chl-aconcentration measurements is verified.

2 Instrumentation

Fig.1 shows that MOFL is composed of three modules,i.e., coaxial transmitter, multi-channel photo-electric conversion and data acquisition module, which are described as follows.

Fig.1 Sketch of MOFL System.

1) The coaxial transmitter module consists of a third harmonic of a Q-switched Nd:YAG laser, a Cassegrain telescope, and high efficiency laser reflectors. The coaxial structure helps adjust backscattering signal reception. Compared with the divergence of laser beams (1 mrad), a Field of View (FOV) (1.2 mrad) is narrow enough to reduce the impact of sunlight during the daytime operation. In addition, a photoelectric trigger is installed near the laser head to control the Analog-to-Digital Converter (ADC) and to monitor fluctuations of laser energy.

2) The multi-channel photo-electric conversion module consists of a spectral splitting structure, a MPMT, and a pulse signal amplifying circuit. In the spectrometer, a diffraction grating with 600 lines/mm is used to separate light of different wavelengths in the received signal, and a spectrum (380–690 nm) is imaged on MPMT with a spectral resolution of 10 nm per channel by a concave mirror. All the receiving channels selected by the Principal Component Analysis (PCA) method (Liet al., 2004) correspond to the characteristic wavelengths of oil and Chl-afluorescence. The center wavelengths of all channels are listed in Table 1. After photons are converted into electrons by MPMT, the pulse amplifier circuit discriminates pulses from direct-current (DC) signals and outputs highgain voltage pulses, which could improve the signal to noise ratio (SNR) of the system (Zhaoet al., 2011). With an approximation that solar background radiation is a DC signal, the capacitance plays a role of the sunlight blocking filter in the pulse amplifier circuit according to the electronic theory (Aiméet al., 2008).

3) The data acquisition module consists of high-speed ADCs, a trigger circuit, and a computer. The electrical signals from MPMT are converted into digital signals which will be saved in the computer by ADCs. In this module, a Visual C++ 6.0 based control software is a core element. Acquisition parameters, such as trigger time, data storage path, and data size, can be preset through the software interface.

The specifications of MOFL are given in Table 1. Table 1 Characteristics of MOFL

In addition to the information of oil spills and Chl-aconcentration, the concentration of dissolved organic matters (DOM) and green terrestrial plants can be detected from a wide range of spectra received by MOFL.

3 Methodology

In this section, the methods to extract oil fluorescence spectra and estimate film thickness (Hoge and Swift, 1980) are described in detail, and the method to obtain the relative concentration of Chl-ais also given in a brief description (Bristowet al., 1981; Hogeet al., 2005).

3.1 Fluorescence of Oil Film

Assuming that the attenuation of laser and fluorescence through the atmosphere is not taken into consideration and the water is deep enough to attenuate the energy of excitation light, the signal received from sea water covered with an oil film of thicknessdconsists of three components,i.e., oil fluorescence (Φoil), the fluorescence of background water (wΦ′) and water Raman scattering (R′), as expressed in the following equation (Hoge and Swift, 1980).

with the subscriptsi,eandrreferring to the components of returned signals at the fluorescence wavelengthλi, the excitation wavelengthλeand the Raman scattering wavelengthλr, respectively.P0is the incident laser power,ψthe Raman conversion efficiency of seawater andδirthe delta function which is equal to 1 in the case ofi=r.ηiandζidenotes the fluorescence conversion efficiency at the wavelengthλifor an optically thick oil film and an optically thick seawater column.keandkiare the extinction coefficients of oil at the excitation wavelengthλeand the fluorescence wavelengthλirespectively. The term exp [-(ke+ki)d] expresses the depression of the excitation laser and the light emission from oil and water at wavelengthλithrough a homogeneous oil film.

According to Eq. (1), the return signal of oil-free water at the Raman channel withd=0 can be given:

where the water fluorescenceΦw=ζrP0, and the water RamanR=ψP0. Therefore, the ratio ofR′/R=exp[-(ke+kr)d] yields:

Using Eq. (3) the film thickness can be calculated. The water Raman is a distinct narrow peak at 404 nm (Hengsterman and Reuter, 1990). As the fluorescence spectral shapes of oil and water are smooth, the Raman intensity at 404 nm can be interpolated (Kung and Itzkan, 1976). Thus, the water Raman intensity can be calculated by subtracting the fluorescence from the received spectra. An extinction coefficient,ke+ki= 0.7 μm-1, measured in laboratory is used to obtain the film thickness.

To obtain the oil fluorescence (Φoil), the background fluorescenceand the water Raman scattering (R′) must be excluded. Herein,Φw,RandR′ are used as the parameters for estimating the background spectrum under two assumptions. First, the extinction coefficients of oil film in observed wavelength ranges are the same. Second, the background water under oil film is the same as the nearby water so that the fluorescence spectral shapes of the waters are the same. Thereforesignal from the nearby waterΦw+Rcan be measured by Lidar.RandR′ can be obtained through interpolation. Thus, the background fluorescence,can be calculated. Based onΦoil, the fluorescence spectrum of the oil film is used to identify oil groups.

3.2 Fluorescence of Chlorophyll-a

In order to evaluate the chlorophyll concentration, a simple two-band ratio algorithm is generally expressed as (Bristowet al., 1981):

wherePFis the detected chlorophyll fluorescence intensity.Ris the detected water Raman backscatter intensity.nfdenotes the chlorophyll concentration.kF,kLandkRare the attenuation coefficients at fluorescence, incident laser, and water Raman wavelengths, respectively.δdenotes the calibrated system parameter.

With the assumption that particulate and dissolved material vary only in concentration but not in spectral absorption characteristics,kF,kLandkRcould be considered as constants. Thus, normalized by the water Raman intensity, the intensity of chlorophyll fluorescence is given as:

whereCis a proportionality constant which can be obtained by making instantaneous ground truth measurements fornfat a small number of reference sites during each field experiment (Bristowet al., 1981). According to Eq. (5), the relative concentration of Chl-acan be given asPF/R.

4 Experiments and Discussion

The data showed in this paper were collected in two field experiments,i.e., an experiment carried out at a 110-meter platform in August 2009 (Liet al., 2010) and an airborne survey over the Yellow Sea in October 2009. In the first experiment, artificial oil films floating over coastal waters were detected by MOFL. The potential contaminations from the films were prevented by oil containment booms in the area. In the second experiment, the fluorescence of Chl-ain the upper layer of the ocean was continuously monitored by MOFL, and the distributions of relative Chl-aconcentrations were mapped and compared with those obtained from the Moderate Resolution Imaging Spectroradiometer (MODIS).

On 26 August, 2009, a series of samples were investigated in the offshore area of Qingdao, located at 36?02? 2.5??N, 120?20?24.1??E. The three-dimensional spectrum of water fluorescence measured by MOFL is shown in Fig.2. The two peaks at close range are the feedback of laser wavelengths at 355 nm and 532 nm. The intensity of the excitation wavelength at 355 nm is used to normalize induced fluorescence and to eliminate fluctuations of laser energy. The light at 532 nm is completely filtered out in the near-field of telescope and is only used to activate the trigger circuit. The emission fluorescence of water is backscattered from a water surface layer. This paper focuses on spectral features to classify oil spills, because the frequency of MOFL (60 MHz) is too low to estimate short fluorescence lifetimes of oils (less than 20 nsec) (Patsayevaet al., 2000). In this experiment, seawater, wakame, phoenix tree, and oils of different types are induced to emit fluorescence, and the area-normalized fluorescence spectra are displayed in Fig.3. Most of the oil fluorescence spectral features are shown in the wavelength range of 400–600 nm, while the fluorescence intensity of clean water tends to decrease from 500 nm tolonger wavelengths (Camagniet al., 1991).

Fig.3 shows that the Raman scattering by seawater is strong enough to distinguish water from other sampling types. Both fluorescence spectra of wakame and phoenix tree show fluorescence peaks of Chl-aat 685 nm (Hogeet al., 1983), but are different in the spectral band from 450 nm to 650 nm. Fig.3(c) shows that different oil types correspond to different fluorescence spectral characteristics in the wavelength range of 400–600 nm. Except for the fuel oil #3F the oils have fluorescence peaks at 495 nm. Yet, castor oil, gasoline #93, and diesel fuel also show sub-peaks between 400 nm and 460 nm, particularly, a distinctive sub-peak at 445 nm associated with the diesel fuel. The heavy oils,i.e., the fuel oil #3F and the crude oil #1C, have fluorescence spectra with only one characteristic peak between 500 nm and 520 nm.

Fig.2 A three-dimensional fluorescence spectrum of water received by MOFL.

Fig.3 Emission spectra of seawater, wakame, phoenix tree and different oil types acquired by MOFL on 27 August, 2009.

To calculate the thickness of oil film, artificial films floating in the seawater off the coast were monitored by Lidar that was deployed in the same way on the platform as in the previous experiment. Owing to low viscosities and poor volatilities, diesel fuel #8 and crude oil #1C were selected as the film material. A small oil slick was created by spilling 0.05 liter of crude oil 1C# on the sea surface. Another slick was created by spilling 0.03 liter of diesel fuel 8#. To evaluate the accuracy of thickness measurements by MOFL, a film thickness, estimated by the volume of oil and the area of the film in the filmmaking, was used as the reference value.

The films and the surrounding waters could be induced to emit fluorescence by adjusting the scanning points of Lidar. Water surface was sampled at 45 points. Detected by MOFL, the profiles of original signal return at typical wavelengths,i.e., 405 nm (the Raman scattering), 445 nm (the laser-induced fluorescence of CDOM), 495 nm and 505 nm (the fluorescence peaks of oils), and 685 nm (the fluorescence emission of chlorophyll), are shown in Fig.4. The intensities of channels changed as different probing points were scanned. Fig.4(f) presents the intensity ratio of the fluorescence channel at 495 nm to the Raman channel at 405 nm (I495/I405).I495/I405at OP-39～OP-45 are larger than the others. The values at OP-35, OP-40 and OP-45, are 0.93, 3.25 and 3.33, respectively. Considering that the original laser-induced fluorescence at OP-39～ OP-43 is weak, it is believed that films are floating on the water surface at OP-44 and OP-45. OP-05, OP-35, OP-40 and OP-45 correspond to the diesel film, seawater, reefand crude oil film, respectively. The fluorescence spectra at these points (OP-05, OP-35, OP-40 and OP-45) are plotted in Fig.5. It is obvious that the fluorescence induced from the reef is extremely weak (Fig.5(c)). As shown in Fig.5(a), the fluorescence from the diesel film results in the water spectrum deviated from the normal state at OP-05. The spectrum is a mixture of the fluorescence of diesel fuel #8 and seawater. However, in terms ofI495/I405, the diesel film at OP-05 cannot be distinguished from others. Thus, both the spectral shape of fluorescence and the value ofI495/I405are essential to determine the distribution of oil spills.

Fig.4 Received Lidar signals in typical channels by scanning the sea surface covered with artificial films. Channel intensities and two-channel ratio (I495/I405) fluctuate at different observed points around the films. Hereafter, the‘observed point’ is abbreviated as ‘OP’ and the fifth observed point is written as ‘OP-05’.

Fig.5 Fluorescence spectra associated with (a) diesel film, (b) seawater, (c) reef and (d) crude film at four observed points (OP-05, OP-35, OP-40 and OP-45).

Used to estimate the film thickness and remove background fluorescence, the intensity of water Raman attenuated by oil films could be easily obtained at the observational points and the background fluorescence could be removed with reference to the fluorescence spectrum of seawater near the film,i.e., the spectrum at OP-35.According to Eq. (3), the film thickness could be obtained by calculating the attenuation of water Raman scattering with the extinction coefficientke+kr=0.7 μm-1. Figs.6(a) and 6(b) show the water Raman intensities and the calculated film thicknesses, respectively. The film thickness at OP-05 is 1.98 μm, larger than the reference value of 1.20 μm. This can be attributed to the angles between the laser beam and the film surface, and the changes of water turbidity in measurements (Karpiczet al., 2006). The film thicknesses at OP-39～OP-45 can not be calculated because water Raman has the zero intensity. Although the intensity of Raman channel (at 405 nm) at OP-45 is greater than zero (Fig.4(a)), the water Raman scattering is actually exhausted by the optical attenuation of crude films and the Raman intensity is zero as shown in Fig.6(a).

Fig.6 (a) Water Raman intensities interpolated at 405 nm based on the intensity of adjacent channels; (b) Calculated film thicknesses (Eq.(3)).

To evaluate the airborne performance of MOFL, the flight experiment was conducted over the Yellow Sea during the daytime of 16 October, 2009. In this experiment, fluorescence signals in the upper layer of the ocean were continuously recorded along a flight path. According to Eq. (5), the intensity of fluorescence at 685 nm can be used to obtain the relative concentration of Chl-a. Fig.7 gives a synoptic view of the chlorophyll distribution calibrated by monitoring laser energy fluctuations in five sections on the flight path. The measurement gaps on the flight path are due to temporary closures of the aircraft observation window. Fig.7 shows that the relative concentration of Chl-ain the offshore waters is higher than 0.7, while it is lower than 0.5 in the open ocean. The changes in Chl-aconcentration demonstrate that the growth of phytoplankton is promoted by high concentrations of nutrients in coastal areas (Babichenko, 2008).

In order to validate the results of the flight experiment, the relative concentrations of Chl-afrom MOFL are compared with MODIS products (NASA database website, http://oceancolor.gs fc.nasa.gov/seadas/). To properly compare with the 1-km resolution MODIS data, the Lidar data are averaged over 75 laser pulses. Fig.8 shows a linear regression line (y=0.99x+0.26) to fit the MOFL and MODIS data. The regression bias (0.26) is mainly due to the different inherent properties of Lidar and MODIS. The root mean square error (RMSE) and the correlation coefficient in the relative concentrations of Chl-abetween MOFL and MODIS are 0.06 and 0.82, respectively. In general, the measurements by MOFL are in agreement with those by MODIS.

Fig.7 The relative concentration of Chl-a recorded by MOFL in the flight experiment. The upper panel shows the fluctuation of the relative concentration of Chl-a on the flight path, and the lower panel reveals the distribution of the relative concentration of Chl-a along the flight path.

Fig.8 Comparison of the relative concentrations of Chl-a from MOFL and from MODIS by linear regression.

5 Conclusions

The MOFL system equipped with a compact multichannel photomultiplier tube (MPMT) is applied to measure the oil spills and relative Chl-aconcentrations in the upper layer of the ocean. In the coastal experiment, oil films, seawater, wakame, and phoenix tree have been identified by analyzing fluorescence spectra. The thickness of oil films can be estimated from the attenuation of water Raman scattering by means of the optical attenuation ofthe oil films. By monitoring artificial films floating over coastal waters it is revealed that the fluorescence spectra induced from surface water and the intensity ratio of two channels (I495/I405) are essential for the discovery of oils slicks. In the flight experiment, MOFL is applied to measure the relative Chl-a concentrations in the upper layer of the ocean. The data comparison indicates that the MOFL and the MODIS measurements have a good agreement on the distributions of Chl-a. The MOFL results are proved to be valid for the fluorescence study.

Even though an approximate film thickness can be computed by Eq. (3), the assumptions that the water covered by oil films is the same as the surrounding water and the attenuation coefficients of different oil types are the same, could result in estimation errors of film thickness (Barbaro et al., 1991; Brown et al., 2003b). Furthermore, the Lidar fluorosensors measurements should be improved in temporal resolution. More investigations are needed to develop airborne or even spaceborne sensors for the monitoring of marine environment.

Acknowledgements

This work was supported by the National High Technology Research and Development Program (2006AA 06Z415) and the Global Change Research Program of China (2012CB955603). The authors wish to express their thanks to the team members of the MOFL of Ocean Remote Sensing Institute, Ocean University of China.

Aimé, L. E., Giuseppe, V., and Amerigo, T., 2008. Automatic gain control for Raman Lidar signals. In: Proceedings of the SPIE 7153, 12/2008, DOI: 10.1117/12.804864.

Babichenko, S., 2008. Laser remote sensing of the european marine environment: LIF technology and applications. In: Remote Sensing of the European Seas. Barale, V., and Gade, M., eds., Springer, Netherlands, 189-204. DOI: 10.1007/978-1-4020-6772-3_15.

Barbaro, A., Cecchi, G., and Mazzinghi, P., 1991. Oil UV extinction coefficient measurement using a standard spectrophotometer. Applied Optics, 30 (7): 852-857.

Bristow, M., Nielsen, D., Bundy, D., and Furtek, R., 1981. Use of water Raman emission to correct airborne laser fluorosensor data for effects of water optical attenuation. Applied Optics, 20 (17): 2889-2906.

Brown, C. E., and Fingas, M. F., 2003a. Review of the development of laser fluorosensors for oil spill application. Marine Pollution Bulletin, 47 (9-12): 477-484, DOI: 10.1016/S0025-326X (03)00213-3.

Brown, C. E., and Fingas, M. F., 2003b. Development of airborne oil thickness measurements. Marine Pollution Bulletin, 47 (9-12): 485-492.

Camagni, P., Colombo, A., Koechler, C., Omenetto, N., Qi, P., and Rossi, G., 1991. Fluorescence response of mineral oils: spectral yield vs absorption and decay time. Applied Optics, 30 (1): 26-35.

Hengsterman, T., and Reuter, R., 1990. Lidar fluorosensing of mineral oil spill on the sea surface. Applied Optics, 29 (22): 3218-3227.

Hoge, F. E., and Swift, R. N., 1980. Oil film thickness measurement using airborne laser-induced water Raman backscatter. Applied Optics, 19 (19): 3269-3281.

Hoge, F. E., Lyon, P. E., Wright, C. W., Swift, R. N., and Yungel, J. K., 2005. Chlorophyll biomass in the global ocean: Airborne Lidar retrieval using fluorescence of both chlorophyll and chromophic dissolved organic matter. Applied Optics, 44 (14): 2857-2862.

Hoge, F. E., Swift, R. N., and Yungel, J. K., 1983. Feasibility of airborne detection of laser-induced fluorescence emissions from green terrestrial plants. Applied Optics, 22 (19): 2991-3000.

Jha, M. N., Levy, J., and Gao, Y., 2008. Advances in remote sensing for oil spill disaster management: State-of-the-art sensors technology for oil spill surveillance. Sensors, 8 (1): 236-255.

Karpicz, R., Dementjev, A., Kuprionis, Z., Pakalnis, S., Westphal, R., Reuter, R., and Gulbinas, V., 2006. Oil spill fluorosensing Lidar for inclined onshore or shipboard operation. Applied Optics, 45 (25): 6620-6625.

Kung, R. T., and Itzkan, I., 1976. Absolute oil fluorescence conversion efficiency. Applied Optics, 15 (2): 409-415.

Li, J., Fuller, S., Cattle, J., Way, C. P., and Hibbert, D. B., 2004. Matching fluorescence spectra of oil spills with spectra from suspect sources. Analytica Chimica Acta, 514 (1): 51-56.

Li, X. L., Zhao, C. F., Qi, M. J., Ma, Y. J., Li, Z. G., and Liu, Z. S., 2010. The experiment of multi-channel oceanographic laser remote sensing of oil spills system at high platform. Periodical of Ocean University of China, 40 (8): 145-150 (in Chinese with English abstract).

Liu, Z. S., Ma, S., Wang, X., and Li, Z. G., 2008. Field detection of chlorophyll-a concentration in the sea surface layer by an airborne oceanographic lidar. Journal of Ocean University of China, 7 (1): 108-112.

Patsayeva, S., Yuhakov, V., Varlamov, V., Barbini, R., Fantoni, R., Frassanito, C., and Palucci, A., 2000. Laser spectroscopy of mineral oils on the water surface. In: Proceedings of EARSeL-SIG-Workshop LIDAR, Dresden, Germany, 6/2000: 106-114.

Utkin, A. B., Lavrov, A., and Vilar, R., 2011. Evaluation of water pollution by LIF LIDAR. In: 5th EARSeL Workshop on Remote Sensing of the Coastal Zone, Prague, Czech Republic, 6/2011: 104-113.

Vasilescu, J., Onciu, T., Jugaru, L., and Belegante, L., 2009. Remote estimation of fluorescence marine components distribution. Romanian Reports in Physics, 61 (3): 721-729.

Yarovenko, N. V., Cruz, M. C. M., Vilas, L. G., Pereira, D., Martín, J., and Palenzuela, J. M. T., 2011. Oil pollution using shipborne LIF/LIDAR. In: 5thEARSeL Workshop on Remote Sensing of the Coastal Zone, Prague, Czech Republic, 6/2011: 94-103.

Zhao, C. F., Li, X. L., and Ma, Y. J., 2011. Multi-channel ocean fluorescence lidar system for oil spill monitoring. Infrared and Laser Engineering, 40 (7): 1263-1269 (in Chinese).

(Edited by Xie Jun)

(Received July 10, 2012; revised September 3, 2012; accepted April 15, 2014)

? Ocean University of China, Science Press and Springer-Verlag Berlin Heidelberg 2014

* Corresponding author. Tel: 0086-532-82032021

E-mail: zhaocf@ouc.edu.cn