Shipborne variable-FOV, dual-wavelength, polarized ocean lidar: design and measurements in the Western Pacific

Qi Liu; Songhua Wu; Songhua Wu; Songhua Wu; Bingyi Liu; Bingyi Liu; Bingyi Liu; Jintao Liu; Kailin Zhang; Guangyao Dai; Junwu Tang; Junwu Tang; Ge Chen; Ge Chen; Ge Chen

doi:10.1364/OE.449554

1. Introduction

The goal of ocean remote sensing is near-real time estimation of all pertinent parameters with the required accuracy. Tremendous efforts have been devoted to testing new instrument and deriving new variables for future space-borne measurements. The parameters and elements in oceanography are vertically stratified with depth, such as temperature, ocean currents, ocean organisms, and particles, which require depth-dependent observations. The traditional ocean optical remote sensing (also known as watercolor remote sensing) is only the integral imaging observation of the near-surface ocean, which cannot meet the requirements of three-dimensional detection for ocean scientific and operational application. Despite the exorbitant price and the sparse sampling, the in-situ ocean equipment is still one of the practical ways for vertical ocean observation. It is necessary to improve the ability of vertical ocean measurement.

Ocean lidar can penetrate the seawater to obtain the high accuracy vertical profile of multi-parameters in the ocean [1,2]. As an active and direct detection technology, the retrieval algorithm of lidar is less dependent on the geophysical empirical model and has higher sensitivity and lower measurement error. Therefore, the development of active optical remote sensing technology based on ocean lidar to achieve all day-night, real-time, continuous three-dimensional ocean detection has become a promising method of ocean color. Since the first bathymetry lidar came out in 1968, different ocean lidars had been developed to detect various parameters and elements in the ocean. Especially the polarized lidar, which is widely used in ocean detection, has better advantages to provide multiple ocean parameters. Based on the elastic Mie backscattering signal, the seawater optical properties can be estimated by retrieving the lidar attenuation coefficient K_lidar over the laser penetrating depth. With the depolarization effect of non-spherical particles on the incident light, the polarized ocean lidar also can recognize the ocean communities by the linear depolarization ratio. In the past ten years, the polarized ocean lidar had developed various applications in ocean research. Vasilkov et al. used an airborne polarized lidar to provide the scattering coefficient profile [3]. Churnside et al. derived a radiative transfer equation for airborne polarized lidar return, detecting scatter layers, fish schools, seawater optical properties, and internal waves [4–6]. Collister et al. designed a ship-borne lidar to explore the combined effects of particle composition and seawater multiple scattering based on the lidar linear depolarization ratio [7,8]. Behrenfeld et al. quantified the phytoplankton biomass and diel vertical migration using the particulate backscattering coefficient b_bp532nm and diffuse attenuation coefficient K_d532nm determined from space-borne polarized lidar CALIOP [9,10]. Chen et al. observed the vertical distribution of subsurface phytoplankton layer in South China Sea using dual-wavelength airborne polarized lidar [11]. Chen et al. reported the planned “Guanlan” ocean remote sensing mission, which comprises a near-nadir pointing oceanic lidar and a dual-frequency (Ku and Ka) interferometric altimetry [12]. The oceanic lidar payload is expected to partially reveal the marine food chain and ecosystem with 10 m vertical interval in the euphotic layer, moving a significant step down to the oceanic mixed layer both dynamically and bio-optically.

In contrast to the successful applications for the atmosphere (clouds [13,14], aerosols [15,16], and dust [17–19]), polarized ocean lidar faces more complicated challenges. The critical difficulty is the strong attenuation of seawater preventing the laser beam from propagating deep water. For the 532 nm laser propagating in offshore seawater, the laser energy attenuation can reach nearly five orders of magnitude when the laser penetrates over the depth of 30 m. The strong backscattered signal of the sea surface and near-surface seawater can easily cause the severe saturation of the photodetector, which causes the nonlinear effect and the after-pulse effect of the photodetector [20], leading to the incorrect inversion results. In addition, the complex ocean environment and internal compositions make it more difficult for lidar to obtain the linear depolarization ratio accurately. In previous studies, benefiting from the technology maturity of the solid-state Nd: YAG laser, the polarized ocean lidar commonly use 532 nm as the detection wavelength. From the nearshore seawater to the open ocean, the seawater qualities are varying due to the concentration of various compositions, leading to the different optimum optical penetration wavelength [21,22]. Furthermore, the lidar detection based on a single wavelength cannot reveal the impact of particle size and absorption on optical parameters. It is necessary and promising to explore the characteristics of various wavelengths propagating in seawater. Using the multi-wavelength volume linear depolarization ratio (VDR) and color ratio of lidar, it is possible to identify and distinguish specific matters in seawater. Meanwhile, the in-situ measurement is essential to validate the optical properties retrieved from ocean lidar measurements and to establish the simulation tool for the space-borne ocean lidar mission in future.

In order to obtain the optical parameters profile and to explore the correlation between the ocean compositions and the lidar VDR and color ratio, the self-developed ship-borne variable-FOV, dual-wavelength, polarized ocean lidar system, LOOP (Lidar for Ocean Optics Profiler), will be introduced in detail in this study. The measurements and retrieval algorithm were validated with in-situ optical instruments in the western Pacific campaign in 2019. Benefitting from the photon-counting detection with long time accumulation at a fixed station, the depth of inversion results at 355 nm and 532 nm, which is determined by signal to noise ratio SNR = 3, reached 30 m and 50 m, respectively. Considering the after-pulse effect of the LOOP system, the results of lidar attenuation coefficient K_lidar in the narrow field of view (FOV) and wide FOV were consistent with the seawater beam attenuation coefficient c, which is the inherent optical parameter of seawater (IOP), and down-welling diffuse attenuation coefficient K_d, which is the apparent optical parameter of seawater (AOP). After correcting of the cross-talk between receiving channels of the LOOP system, the measured VDR profile and Chl_a concentration profile showed a good correlation. By comparing the attenuation and polarization characteristics of 355 nm and 532 nm in seawater, the color ratio based on the dual-wavelength of K_lidar is consistent with the CDOM concentration profile. The combination of the K_lidar, the VDR and the color ratio provides the possibility for dual-wavelength ocean lidar, such as the ultraviolet and visible wavelength, to resolve the characterization of layers in the subsurface ocean.

2. LOOP system

The optical layout of the LOOP system is presented in Fig. 1. The LOOP is mainly composed of four subsystems: the transmitting subsystem, the receiver and spectroscopy subsystem, the detector and recorder subsystem, and the controller and auxiliary subsystem. The first three subsystems will be introduced in detail in the following sections. The controller and auxiliary subsystem mainly consist of electronic device including the real-time network transmission devices, the monitoring computers, the drying and the heat dissipation devices dedicating to marine hash environment.

Fig. 1. The optical layout of the LOOP system.

Download Full Size | PDF

2.1 Transmitting subsystem

Sea water has tremendous absorption at near-ultraviolet and near-infrared wavelength, leaving a solar irradiance in “blue-green window” for aquatic life. The wavelength of maximum transmittance shifts from blue in the cleanest open ocean water to green in the chlorophyll-rich coastal water. The 532 nm wavelength, the second harmonic frequency of Nd:YAG solid state laser, is within the optimal practicable optical window for most coastal seawater areas with high concentration of suspended material and chromophoric dissolved organic matter (CDOM) [21]. While the optimal wavelength will shift to near 450 nm, when the beam attenuation coefficient c of the seawater is closing to the pure seawater. The “Guanlan” mission team propose a design using 486.13 nm wavelength generated by the 355 nm laser pumped optical parametric oscillator for clean open ocean. For a shipborne lidar system, the hash marine environment of high salt and humidity, as well as the vibration and sloshing of the ship, make the laser source of the LOOP system confronting great challenge. Considering the reliable laser source with compact structure and high peak energy output, a commercially mature and compact Nd: YAG laser capable of emitting lasers at 532 nm and 355 nm is adopted for the LOOP system. The 355 nm is dedicated to excite the fluorescence signal of seawater in detection of the Chl_a and CDOM. Benefitting from the dual-wavelength design, the LOOP system can explore the elastic scattering characteristics of particles at two wavelengths in the clean seawater, such as the K_lidar, VDR, lidar ratio and color ratio.

As shown in Table 1, the transmitter of the LOOP system used the second and third harmonic frequency of compact high peak power flash lamp-pumped Nd: YAG pulses laser, Quantel Q-smart 450, which emits 120 mJ and 200 mJ output energy at wavelengths of 355 nm and532 nm, respectively. With two wavelengths of laser transmitted in one beam, the laser pulse duration is 5 nanoseconds at a 20 Hz repetition, and the linear polarization degree is greater than 100:1. Both 355 nm and 532 nm lasers are used to excite the elastic scattering of seawater to obtain VDR and K_lidar. In addition, the 355 nm laser is also capable of exciting seawater Raman (404 nm) and fluorescence (680 nm) signals, which can be used to inverse the Chl_a concentration for further research. The transmitting laser is not expanded or collimated, and the divergence angle is less than 0.5 mrad. The front-end window mirror of the LOOP system uses a high-transmittance quartz glass coated with anti-reflection film to prevent laser reflection from damaging the internal devices.

Table 1. Parameters of transmitting subsystem

View Table | View all tables in this article

2.2 Receiver and spectroscopy

The specifications of the receiver and spectroscopy subsystem are shown in Table 2. The receiver telescope is a 60 mm diameter aspheric lens with an iris diaphragm placed at the focal point, which enables the FOV to be adjusted continuously from 16 mrad to 60 mrad. Since the system is installed at the ship side and is close to the sea surface, the strong near-field signal caused by the backscattering from the sea surface, near-surface seawater (usually within 3 m depth), ocean spray, and droplets, will easily saturate the PMT detector, so the system FOV sets to appropriate overlap factor to mitigate the saturation effect. To achieve this, the receiver optical axis of the lens is set biaxial to the laser optical axis by 350 mm, and the laser transmitting direction is biased towards the receiver optical axis direction by 8 mrad. As a result, the FOV of the system reaches full overlapping at the distance of 21.9 m and 9.7 m, when the receiving FOV is set as 16 mrad and 60 mrad, respectively.

Table 2. Parameters of receiver and spectroscopy subsystem

View Table | View all tables in this article

Table 3. Parameters of detector and recorder subsystem

View Table | View all tables in this article

After focused by the aspheric lens, the receiving backscattered photons are first collimated by a cemented achromatic doublet and transmits into the spectroscopy subsystem. Then the photons are separated to 532 nm, 355 nm, 404 nm and 680 nm in sequence according to the notch filter and two long-wave-pass filters which are arranged in a tandem array. Citing the “Δ90°-calibration” method [16,23], a zero-order half-wave plate, which is design to calibrate the depolarization ratio of the LOOP system, is placed in a continuous rotation mount before the polarization beam splitter (PBS) of 532 nm. Two PBSs are concatenated and separate photons into two orthogonal polarization states, as well as minimize the cross-talk between these two channels. The dual-PBSs design ensures that the extinction ratio of parallel polarization to orthogonal polarization reaches 10⁶: 1. The polarization beam splitting method of 355 nm is similar to that of 532 nm, but the principle axis is opposite to that of the 532 nm channel. At the end of each receiving channel, the photons pass through the narrow band filter and then focus on the photocathode of the photomultiplier tube (PMT) through a cemented achromatic doublet. All the optical components in the spectroscopy subsystem are assembled using uniform standard lens tubes and adapters, so that the mechanical installation error is minimized. The enclosed assembly devices can also eliminate the interference of the stray light inside the system.

2.3 Detector and recorder

The specifications of the detector and recorder subsystem are shown in Table 3. Six PMTs (Hamamatsu H10721-210P), which work in photon counting mode, are used to detect the backscattered photons. The PMT is extremely sensitive to the incident photons, and its detection capacity is restricted by the maximum count rate, which is mainly caused by the “dead time” effect. When the number of incident photons exceeds its maximum count rate, the PMT will fail to respond to all the incident photons. As a result, the number of output photon events will not be linear to the intensity of the incident signal. In order to constrain the intense signal from the sea surface, an absorptive neutral density filter (NDF) is added to each PMT. The NDF’s optical density (OD) of the 355 nm, 532 nm, 404 nm, and 680 nm channels are 2, 5, 4, 3, respectively. The OD values are designated according to the field test and the criterion that the peak current of the PMT is lower than 400 μA, which not exceed the maximum saturation pulse output peak current. Finally, the backscattered signal is recorded by a six-channel digitizer at a sampling rate of 1 GHz for photon counting (PC), which represents the vertical depth resolution in the ocean is 0.11 m.

3. Methodology

This section will introduce the inversion methods of K_lidar and VDR in detail, which are described in Section 3.1 and 3.3, respectively. In order to verify the LOOP system measurements of K_lidar tending to c and K_d with the narrow FOV and the wide FOV, respectively, the method of calculating c profile based on the model which is related to the in-situ measured Chl_a profile will be described in 3.2. The K_d profiles is the in-situ measurement which will be introduced in Section 4.

3.1 K_lidar of the LOOP system

The depth-dependent lidar equation in seawater can be written as [24]

(1)$${P(z )= \frac{{{E_0}AO(z )T_{atm}^2T_{sur}^2{\eta _r}}}{{2n{{(nH + z)}^2}}}{\eta _{QE}}\nu \beta ({\pi ,z} )\exp [ - 2\mathop \smallint \nolimits_0^z {K_{lidar}}({z\mathrm{^{\prime}}} )dz\mathrm{^{\prime}}]}, $$

where P(z) is the received backscattered signal of seawater, E₀ is the energy of the transmitting laser pulse, n is the refractive index of the seawater, O(z) is the lidar overlap function, which is related to the depth z. A is the area of the receiving telescope, H and z are the laser path length in the atmosphere and seawater, respectively. T_atm, T_sur are the transmittance of the atmosphere and sea surface, respectively, η_r and η_QE are the optical efficiency of system and quantum efficiency of the PMT, respectively, ν is the light speed in vacuum, β (π, z) is the volume scattering coefficient at scattering angle π, K_lidar is the lidar attenuation coefficient. Using the quasi-single scattering approximation, the K_lidar in seawater can be described as

(2)$${{K_{lidar}} ={-} \frac{1}{2}\frac{\textrm{d}}{{dz}}ln({P(z ){z^2}} )}. $$

Based on the Monte Carlo Simulations, Gordon proposed that the value of K_lidar is related to the parameters of the lidar system and the optical characteristics of the seawater. When the product of cR, i.e., the lidar foot spot radius on the sea surface R multiply the beam attenuation coefficient of seawater c, is far less than 1, the K_lidar can be approximately equal to c. When the cR is greater than 5 or 6, K_lidar is approximately equal to the diffuse attenuation coefficient K_d [25]. Subsequently, Phillips et al. used the semi Monte Carlo method to improve the theory in 1984 [26]. When the lidar is under the wide FOV, the K_lidar is larger than the absorption coefficient a and smaller than the sum of a and backscattering coefficient b_b, i.e., a < K_lidar < (a + b_b). When the FOV is extremely wide that the cR is much larger than 1, the value of K_lidar approaches to a [27].

3.2 Calculation of Inherent Optical Properties of the seawater

In order to verify the inversion results of the LOOP system, the total seawater absorption coefficient a_total and total seawater backscatter coefficient b_total are calculated by a bio-optical model, then the total seawater beam attenuation coefficient c_total= a_total + b_total is obtained.

The a_total can be expressed as

(3)$${{a_{\textrm{total}}} = {a_w} + {a_p} + {a_g}}, $$

where a_w is the pure seawater absorption coefficient, a_p and a_g are the absorption coefficient of phytoplankton and CDOM, respectively. In the open ocean, the total non-water absorption coefficient a_nw is equal to a_p +a_g. In this study, a_w uses the measurement results of Lee [28]and Mason [29], for the wavelength at 532 nm, i.e. a_w(532) = 0.043 m^-1. Since the trial region in the western Pacific belongs to the clean open ocean, the inherent optical parameters can be estimated by the concentration of Chl_a. Then, a_p is calculated using the method proposed by Bricaud [30] with wavelength correction coefficient at 532nm

(4)$${{a_p}({532} )= 0.0155 < Ch{l_a}{ > ^{0.7985}}}. $$

Meanwhile, a_g is calculated using the empirical equation proposed by Morel and Gentili [31,32]

(5)$${{a_g}({532} )= 0.006 < Ch{l_a}{ > ^{0.63}}}, $$

The calculation of b_total is similar to a_total, and it can be expressed as

(6)$${b_{\textrm{total}}} = {b_w} + {b_p}, $$

where b_w is the pure seawater scatter coefficient, b_p is the scatter coefficient by phytoplankton. The b_w and b_p both can be calculated using the method supposed by Morel and Loisel [33,34]

(7)$${{b_w}({532} )= 0.0028{{\left( {\frac{{532}}{{500}}} \right)}^{4.3}}}, $$

(8)$${{b_p}({532} )= 0.416 < Ch{l_a}{ > ^{0.766}}\left( {\frac{{532}}{{550}}} \right)}. $$

3.2 Volume linear depolarization ratio of LOOP system

Lidar linear depolarization ratio has different definition of VDR and particle linear depolarization ratio. Since various complex factors can cause the lidar depolarization ratio to change during the campaign, the polarization correction of the system is of vital importance. Only the correction method of VDR is introduced in this study. The VDR is defined as the ratio of the orthogonal polarized backscattered signal to the parallel polarized backscattered signal [35]. The accurate measurement of the VDR depends on the determination of the polarization correction coefficient, including the gain ratio of different channels, the misaligned angle of the lidar system, and the cross-talk of the PBS. Based on the “Δ90°-calibration” method [16,23], combining with the specific optical design of the LOOP system, the accurate correction of VDR is completed based on the measurement of the half-wave plate (HWP) at different characteristic angles, i.e. 0° and 45°. The calibrated VDR of the target detected by lidar δ^ν can be expressed as

(9)$$\begin{aligned} {\delta ^\nu } &= \frac{{{P_R}(z ){T_P} - G{P_T}(z ){R_P} + ({{P_R}(z ){T_S} - G{P_T}(z ){R_S}} )ta{n^2}\phi }}{{G{P_T}(z ){R_S} - {P_R}(z ){T_S} + ({G{P_T}(z ){R_P} - {P_R}(z ){T_P}} )ta{n^2}\phi }}\\ &= \frac{{m(z ){T_P} - G{R_P} + ({m(z ){T_S} - G{R_S}} )ta{n^2}\phi }}{{G{R_S} - m(z ){T_S} + ({G{R_P} - m(z ){T_P}} )ta{n^2}\phi }} \end{aligned}, $$

where m(z) is the measured depolarization ratios. G is the electro-optical gain ratio between the mutually orthogonal channels. ϕ is the transmitter-receiver misalignment angle, i.e., the rotation angle of the polarization plane. R_p and R_s are the reflectivity of the parallel and orthogonal polarized backscattering light to the plane of the PBS, respectively. T_p and T_s are the transmittances of the parallel and orthogonal polarized backscattered light to the plane of the PBS, respectively. The parameter G is measured at a high-altitude region in a clean atmosphere, where exists little aerosols. By rotating the half-wave plate to reach the minimum depolarization ratio, the ϕ of the 532 nm channels of the LOOP system is 12 degrees. The G factor of 532 nm channels in the photon counting mode is 9.47. Similarly, the ϕ of the 355 nm channels is 0 degrees, and the G factor is 1.12. The high G factor of 532 nm channels is due to the rotation of the linear polarization direction when the 532 nm laser is emitted through the triplex crystal.

4. Measurements

4.1 Lidar deployment in the western Pacific

The field campaign of the Active and Passive Optics Experiment in the Western Pacific (APOEWP) was organized aboard the research ship from October 23, 2019, to December 10, 2019. The observing stations are shown in Table 5. The trial area is an oligotrophic ocean belongs to the western end of the North Equatorial Circulation and the southern birthplace of the Kuroshio, where the Chl_a concentration observed by MODIS (shown in Fig. 2 left top) was perennially less than 0.06 mg/m³. The absorption of CDOM in the clean open ocean can be small compared to absorption by Chl_a. The LOOP system was installed on the ship’s rail, 13 m above the sea surface with an inclination angle of 15° from nadir to minimize the specular reflection from the sea surface, as well as the foam and splash near the hull (shown in Fig. 2 left bottom). In this campaign, the FOV of the LOOP system was set to two limit values, the narrow FOV of 16 mrad, the wide FOV of 60 mrad. The FOV of the system reached the full overlap at 3.7 m (vertical height) above the sea surface and 7.2 m (vertical depth) below the sea surface with a wide FOV and narrow FOV, respectively.

Fig. 2. The field measurement of the LOOP system with in-situ optical instruments in the western Pacific. The colorbar of the map refers to the Chlorophyll a concentration provided by MODIS.

Download Full Size | PDF

In order to verify the measurements of the LOOP system, a series of in-situ instruments were deployed to provide ocean optical parameter profiles. The Spectral Absorption and Attenuation Sensor ac-s (WET Labs Inc.) was used to measure the a_nw. The Scattering Meter ECO BB-9 (WET Labs Inc.) was used to measure the Chl_a concentration, the total particulate backscattering coefficient (b_bp), and the CDOM. The hyperspectral radiance and irradiance sensors RAMSES (TriOS) was used to measure the K_d. All instruments except the RAMSES were mounted on the ocean optical profile cage, which submerged uniformly from the sea surface at a speed of 0.15 m/s by winch. After counterweighting, the RAMSES was manually placed into the sea with cables, drifting far from the ship to avoid ship shadow and free diving. The instrument information is shown in Table 4.

Table 4. In-situ optical instrument and measurement parameters

View Table | View all tables in this article

4.2 Lidar backscattered signal profile

Since the backscattered signal of the sea surface is always the strongest during the field measurements, the position of the backscattered signal of the sea surface is set as the time-zero point. Due to the narrow FOV and the wide FOV share the same optical path with a single diaphragm, the two FOVs cannot be measured synchronously and were generally measured stepwise. After correcting the lidar incline angle and sea surface refraction angle, the PC sampling frequency of 1 GHz achieved the vertical depth resolution reached 0.11 m in seawater with a refractive index of 1.33. Every 40 pulses (2 seconds) of the backscattered signal are accumulated into a single file in the recorder. After accumulating for 3 hours, the files are accumulated into a single profile, which is the backscattered signal profile of parallel or orthogonal polarization. Then the signal-to-noise ratio (SNR) is calculated to evaluate the effectiveness depth of backscattered signal

(10)$${SNR = \frac{{{P_z}}}{{\sqrt {{P_z} + {P_{bg}}} }}}, $$

where P_bg is the background signal.

4.3 After-pulse effect

In contrast to the atmospheric lidar, the after-pulse effect of PMT should be considered in ocean exploration. After-pulses usually follow a large amplitude pulse, which affect the accurate measurement of the weak signal and cause errors in pulses counting application. There are two types of after-pulses, i.e., the short delay after-pulse (SDAP) and the long delay after-pulse (LDAP). SDAP usually appears several nanoseconds to tens of nanoseconds after the pulse signal, causing a spike-like waveform, which is often mistaken as a backscattered signal of the scattering layer [20]. While the LDAP may appear later to hundreds of nanoseconds to several microseconds, after the primary signal, which will cause a false signal tailing. In the ocean lidar applications, the LDAP is often mistaken as the backscattered signal in the deep seawater, which may overestimate the lidar penetration depth in the seawater [36]. Based on the test experiment in the lab, with controlling the peak current of the PMT, the after-pulse effect of the PMT can be neglected by selecting an appropriate SNR threshold.

In order to evaluate the after-pulse effect of the LOOP system measured in the western Pacific, a hard target experiment is organized. When the laser hits to the wall surface, the peak position of the backscattered signal which is directly reflected by the hard target surface is set as the time-zero point. Using the optical attenuators, the received backscattered signal intensity of the wall surface on PMT is adjusted to be equal to that of the sea surface. After accumulating the same number of bin files as used in the ocean experiment, the wall surface reflected signal profile is given, and the signal after time-zero is considered as the after-pulse signal. Based on the method of deconvolution, the hard target reflected signal profile is used as the transient response function F(z) (shown in Fig. 3). The deconvolution process can be described as

(11)$${{P_\textrm{r}}(z )= F(z )\ast {P_c}(z )}, $$

where P_r(z) is the backscattered signal received by the detector, and P_c(z) is the corrected backscattered signal after deconvolution. Different from the results of Lu and Li, the after-pulse signal intensity is nearly three orders of magnitude weaker than that of the primary signal. Comparing the raw backscattered signal of seawater with the corrected backscattered signal after deconvolution, the attenuation rate is roughly the same, which means that the after-pulse effect of the LOOP system measured in the western Pacific campaign has been corrected.

Fig. 3. A hard target (wall surface) experiment is organized for evaluating the after-pulse effect of the LOOP system measured in the western Pacific campaign. The yellow line corresponds to the backscattered signal of the wall surface, which is used as the transient response function in the deconvolution method. The blue line corresponds to the measured backscattered signal of seawater in the western Pacific, while the red line is the deconvolution result. The peak position of the backscattered signal, which is reflected by the wall surface and sea surface, respectively, is set as the time-zero point, and the time has been converted to the ocean vertical depth.

Download Full Size | PDF

5. Results and discussion

5.1 K_lidar compared with seawater optical property parameters

Based on the theory proposed by Gorden [25] and Walker [27], the K_lidar is equal to c, K_d, a with the cR<<1, 5<cR<6, cR>>1, respectively (R is the foot spot diameter of lidar FOV at sea surface). When the system has a narrow FOV (cR<<1), the backscattered signal is mainly contributed by the single scattering of seawater so that the K_lidar is equal to c. when the FOV increases, the received multiple scattering signal of seawater is enhanced, resulting in the K_lidar value is close to K_d then to a. The results shown in Fig. 4 demonstrate the comparison between the measured backscattered signal (solid blue lines) and the simulated backscattered signal (dotted red lines), which is calculated by the lidar quasi-single scattering equation at Station 4. The β_π and K_{lidar-equation} profiles used in the lidar equation are calculated by the bio-optical model [37–39], which is related to the in-situ measurement of Chl_a profile. Limited to the overlap with the narrow FOV, the backscattered signal within seawater depth of 10 m is not detectable. Within the depth of 25 m with wide FOV and the depth of 50 m with narrow FOV, the measured backscattered signal is consistent with the theoretical result, which indicates that the backscattered signal is mainly contributed by single scattering of seawater. However, below the depth of 25 m with wide FOV, the measured backscattered signal gradually deviates from the theoretical result, which indicates that the multiple scattering effect is enhanced with the depth deeper [40,41].

Fig. 4. Measured backscattered signal of seawater compared with calculation result by the lidar quasi-single scattering equation with the FOV of 60 mrad (a) and 16 mrad (b), respectively. The solid blue line is the backscattered signal of seawater measured by the LOOP system at Station 4. The dotted red line is the simulated backscattered signal, which is calculated by the lidar quasi-single scattering equation. The β_π and K_{lidar-equation} used in the lidar equation are calculated by the bio-optical model with the in-situ measurement of Chl_a profile.

Download Full Size | PDF

The comparison between the K_lidar and the in-situ measurements of seawater optical parameters is shown in Fig. 5. The solid green lines in Figs. 5(a) and (d) correspond to the in-situ measurement of Chl_a concentration at station 2 and 4, respectively. The lack of the Chl_a concentration within depth of 10 m is due to the inaccurate measurement caused by the harsh wind and wave on the sea surface. The solid blue and red lines in Figs. 5(b) and (e) correspond to in-situ measurement of a_nw and b_bp, respectively. The dotted blue and red lines in Figs. 5(b) and (e) correspond to a_total and b_total, which are calculated by the model introduced in Eq. (3) and Eq. (7), respectively. The solid blue and red lines in Figs. 5(c) and (f) correspond to the K_lidar measured by the LOOP system with the FOV of 60 mrad and 16 mard, respectively. The solid yellow lines in Figs. 5(c) and (f) correspond to the in-situ measurement of K_d, and the dotted purple lines correspond to c_total, which is calculated by c_total= a_total + b_total. The a_nw, b_bp, and K_d were measured by the ac-s, BB9, TriOs, respectively.

Fig. 5. The measurement of K_lidar profile and bio-optical parameters at Station 2 and 4. (a) The measured Chl_a concentration profile at Station 2. The lack of the Chl_a concentration within depth of 10 m is due to the inaccurate measurement caused by the harsh wind and wave on the sea surface. (b) The in-situ measured a_nw and b_bp (solid lines) at 532 nm compared with the a_total and b_total (dotted lines) at 532 nm calculated by the bio-optical model related to Chl_a concentration. (c) The comparison between the K_lidar and the seawater optical properties. The solid blue and red lines are the K_lidar measured with the FOV of 60 mrad and 16 mrad, respectively. The yellow line is the K_d profile measured by the in-situ instrument of TriOS. The dotted purple line is the c_total calculated by a_total + b_total. The K_lidar is filtered by a moving average filter, whose window width is 30 ns, and the curve is intercepted when SNR is lower than 3. (d), (e) and (f) are the corresponding results measured at Station 4.

Download Full Size | PDF

Considering the complex and harsh environment caused by sea surface wind and waves, which can induce a large error of backscattered signal, the results measured by lidar and in-situ instruments within the depth of 10 m are ignored. In the clean open ocean, the seawater absorption and scattering coefficient have a minimal value, and the value of K_d is much close to the c_total. As shown in Fig. 5(c) and 5(f), the K_lidar is consistent with the c_total, when the FOV is 16 mrad. However, the K_lidar is close to the K_d, when the FOV increases to 60 mrad, within the depth of 25 m. At Station 4, i.e., in Fig. 5(f), a scattering layer leads to an increase of K_lidar between the depth of 10 m and 25 m, which is consistent with the measured Chl_a concentration profile shown in Fig. 5(d). Due to the enhancement of the multiple scattering effect with the wide FOV below the depth of 25 m, K_lidar attenuates and approaches a_total and a_nw in turn. Generally, these results are consistent with the conclusions of Gorden and Walker, which indicates that the measurement of the LOOP system is reliable.

5.2 Volume linear depolarization ratio of 532 nm

The VDR profiles of 532 nm measured at four stations are shown in Fig. 6. The VDR maintains a stable value within the depth of 25 m and then increases with depth in the seawater, which also indicates that the multiple scattering effect in seawater will enhance below the depth of 25 m. In addition, the VDR profile at Station 4 is generally smaller than that at the other three stations. The possible reason is that the area of Station 1, 2, 3 is much different from Station 4 (Table 5). Although the Chl_a concentration is nearly similar, plankton species and living habits in the seawater may differ between stations 1, 2, 3 and station 4. The shapes of plankton living around the area of station 1, 2, 3 may be close to non-spherical, which results in a high VDR, while the shapes of plankton at station 4 may be close to spherical, which causes VDR to be lower. There are also differences in the VDR between wide and narrow FOV measured at stations 2 and 4. It is generally understood that the VDR with narrow FOV is usually smaller than the VDR with wide FOV because of the weak multiple scattering effect, as is shown in Fig. 6(d). However, the result in Fig. 6(b) is quite different, as the VDR with narrow FOV is close to the VDR with the wide FOV. A possible reason is that the non-spherical plankton at station 2 has more motion active than that at Station 4. As the measurement time with narrow FOV lasted from night to midnight (as shown in Table 5), the plankton may float upward at night, causing the VDR increased. The phenomena may also explain the increase of VDR with narrow FOV below the depth of 25 m at Station 4. The differences in plankton species and living habits in seawater may be the main reason for the differences in the VDR between the two stations. Unfortunately, there is no in-situ sampling for biological analysis in this campaign to verify the conjecture.

Fig. 6. The measured VDR of 532 nm at different stations. The solid blue lines in (a), (b), (c) and (d) correspond to the VDR measured by the wide FOV, at station 1 to 4, respectively. The solid red lines in (b) and (d) correspond to the VDR measured by the narrow FOV at station 2 and 4, respectively. Limited by the time arrangement of navigation task, there is no measurement with the narrow FOV at station 1 and 3. The results are smoothed by a moving average filter with a 15 ns window width, and the curve is intercepted when the SNR is lower than 3.

Download Full Size | PDF

Table 5. Summary of the location and time of the measurements

View Table | View all tables in this article

In the open ocean with scant nutrients, it can be assumed that the suspended particulate matters in seawater is mainly contributed by the phytoplankton, which can be characterized by the VDR profile in the seawater. Figure 7 shows good correlation between the VDR of 532 nm and the Chl_a concentration at Station.4. The solid purple, green and blue lines in Fig. 7(a) correspond to the b_bp532nm, the Chl_a concentration, and the VDR, respectively. It can be easily recognized that the b_bp532nm, the Chl_a concentration, and the VDR profiles are consistent in the scattering layer along with the seawater depth. For the linear fitting of the Chl_a concentration and the VDR, as shown in Fig. 7(b), the determination coefficient R² reaches 0.8.

Fig. 7. The correlation between the VDR of 532 nm and the Chl_a concentration at Station 4. (a) The b_bp532nm, the Chl_a concentration, and the VDR profile under different x-axis coordinate scales. (b) The linear correlation fitting between VDR and Chl_a.

Download Full Size | PDF

The linear fitting of the Chl_a concentration and the VDR at other stations is shown in Fig. 8. The R² at stations 1 and 3 reach 0.96 and 0.91, respectively, demonstrating a good correlation between the Chl_a concentration and the VDR. However, as shown in Fig. 8(b), the fitting curve has a slight fluctuation along with the depth, which is mainly due to the measurement error caused by the harsh environment near the sea surface, i.e., the sea waves, droplets, bubbles, and ship sloshing. Moreover, various factors in deep seawater make the backscattered signal seriously attenuated with low SNR, leading to errors in the correlation analysis. From the fitting results at four stations, it is a promising method to establish the correlation between the VDR of lidar and Chl_a concentration in the clean open ocean, making the ocean polarized lidar capable of observing and evaluating the compositions under seawater.

Fig. 8. The linear correlation fitting between VDR and Chl_a at Station 1, 2, 3, respectively. The low linear correlation at Station 2 is due to the measurement error caused by the harsh environment near the sea surface.

Download Full Size | PDF

The remote sensing product of Chl_a concentration is an important data source for studying the ocean primary productivity and the carbon sequestration capacity of the ocean ecosystem. In order to ensure the data reliability, it is essential to verify the accuracy of the remote sensing products. The Chl_a concentration measured by MODIS, limited by passive optical sensors, is only the integral result of the near-surface ocean. Figure 9 shows the comparison among the Chl_a concentration measured by MODIS, the in-situ instrument, and the LOOP observations with SNR=3 and SNR=1, respectively. The Chl_a concentration of the MODIS measurements is the average of 5×5 grids with 1 km resolution of level2 data centered at the LOOP observation station on the same day. The Chl_a concentration of the in-situ measurements is the integral average from the depth of 10 m to 80 m. The Chl_a concentration by the LOOP observation is the integral average of the Chl_a concentration profile based on the correlation between the VDR and the in-situ measurements of Chl_a concentration using the linear regression shown in Fig. 7(b) and Fig. 8. At four stations, the average of MODIS’s Chl_a concentration is about 0.047 mg/m³, close to the average of the in-situ measurements and the average of LOOP’s Chl_a concentration with SNR=1, while the average of LOOP’s Chl_a concentration with SNR=3 is about 0.028 mg/m³. The result shows that the Chl_a concentration observed by the LOOP system within the depth of 50 m is lower than that measured by MODIS and in-situ measurement within the depth of 80 m. The reason may be that the MODIS’s Chl_a concentration is not only the integral within the depth of 50 m, due to the high transparency of the seawater in the clean open ocean, but from the deeper seawater depth, where the maximum phytoplankton layers exist. If the lidar reference integration depth is based on SNR = 1, which is considered to be the lidar penetration depth in this study, the Chl_a concentration estimated by lidar is roughly equal to the other two methods. This comparison indicates that the results of the ocean lidar profiles can provide evaluation and validation for the traditional ocean color remote sensing measurements.

Fig. 9. The comparison among the Chl_a concentration measured by MODIS (the green columns), the in-situ instrument (the yellow columns), and the LOOP observations with SNR=3 (the blue columns) and SNR=1 (the red columns), respectively. The Chl_a concentration measured by the MODIS is the average of 5×5 grids with 1 km resolution of level2 data centered at the LOOP observation station on the same day. The Chl_a concentration of the in-situ measurement is the integral average from the depth of 10 m to 80 m. The Chl_a concentration of the LOOP observation is the integral average of the Chl_a concentration profile correspond to the correlation shown in Fig. 8.

Download Full Size | PDF

5.3 Comparison of measurement results of 355 nm and 532 nm

The comparison experiment on synchronous measurement of 355 nm and 532 nm with wide FOV was organized at stations 1, 2, 3 (Fig. 10). The wavelength of 355 nm has a smaller absorption coefficient and larger scattering coefficient than that of 532 nm in pure seawater, which indicates that, as shown in Fig. 10(a), (d), and (g), the backscattered signal attenuation of 355 nm parallel polarization channel near the sea surface is lower than that of 532 nm parallel polarization channel. The strong seawater backscattering of 355 nm leads to a high value of the VDR at the sea surface and then decreases within the depth of 5 m (Fig. 10(c), (f), (i)). Along with the depth increasing, the backscattered signal of 355 nm decays rapidly, which proves that the multiple scattering effect of 355 nm with wide FOV is stronger than that of 532 nm, leading to a larger K_lidar. Below the depth of 18 m, the K_lidar of 355 nm and 532 nm tend to be the same, which is different from the common knowledge of ocean color optics. In terms of definition, the K_lidar is not a parameter of seawater optical properties, i.e., IOPs/AOPs, but a parameter which describes the attenuation of the received backscattered signal related to the lidar FOV, especially for the multiple scattering effect in seawater with wide FOV. For the ocean lidar to detect the seawater optical properties, K_lidar is just approximating IOPs/AOPs in numerical value. The K_lidar is a function of the wavelength, platform altitude, detection range, receiver FOV, laser beam width and divergence, as well as the seawater optical properties and hydrosol scattering phase function. In the field experimental, the K_lidar also can easily show different performance affected by the laser incident angle, droplets and bubbles caused by harsh sea surface waves, multiple scattering effect of various seawater qualities.

Fig. 10. The comparison of results of 355 nm and 532 nm at Station 1, 2, 3. (a), (b), (c) show the comparison of backscattered signal, K_lidar, VDR, respectively, at Station 1. The corner marker p in the legend of (a) represents the parallel channel, whereas o represents the orthogonal channel. (d), (e), (f) show the results at Station 2. (g), (h), (i) show the results at Station 3.

Download Full Size | PDF

Figure 10(c) shows that the VDR of 355 nm at Station.1 is smaller than that of the other two stations. Based on the lidar measurement time shown in Table 5, the reason may be that the VDR differs in wavelengths and varies with periods due to the plankton or physical field environment changes in the seawater. Unfortunately, because of the power supply failure caused by ship vibration during the navigation, there is no data of 355 nm at the Station. 4. The propagation characteristics of 355 nm and 532 nm in seawater cannot be compared with different seawater conditions. After the campaign introduced in this study, the LOOP system organized a new voyage in the South China Sea, obtained more complete data of backscattered signal of 355 nm and 532 nm and the in-situ data of seawater optical parameters. In follow-up studies, we expect a detailed comparative analysis.

The VDR can only provide the size shape information of ocean particles. With the combination of the color ratio and the VDR, it is expected to realize the identifying of the ocean compositions classification. The color ratio, which is expressed as the VDR color ratio and the K_lidar color ratio in this study, is retrieved based on the backscattered signals of 355 nm and 532 nm. The VDR color ratio is expressed as

(12)$${ColorRati{o_{VDR}} = \frac{{VD{R_{355\textrm{nm}}}}}{{VD{R_{532\textrm{nm}}}}}}, $$

and the K_lidar color ratio is expressed as

(13)$${ColorRati{o_{{K_{lidar}}}} = \frac{{{K_{lidar}}_{355\textrm{nm}}}}{{{K_{lidar}}_{532\textrm{nm}}}}}. $$

Figure 11 presents the comparison of the LOOP measurements of the VDR color ratio and the K_lidar color ratio with the in-situ measurements of the CDOM and Chl_a concentration at stations 1, 2, 3. The blue, red, yellow, and green lines refer to the VDR color ratio, the K_lidar color ratio, the CDOM concentration, and the Chl_a concentration, respectively. The solid, dashed, and dotted lines refer to the measurements at stations 1, 2, and 3, respectively. In the clean open ocean, the optical attenuation is mainly caused by the seawater molecules and phytoplankton. Since the Chl_a concentrations behave similar values and trends at the three stations (Fig. 11(d)), the K_lidar of the 355 nm and 532 nm at the three stations, respectively, have little differences (Fig. 10(b, e, h)). Only based on the Chl_a concentration and the K_lidar, the biological differences among the three stations are hard to be found. However, Fig. 10(c, f, i) show slight differences in the VDR of 355 nm and 532 nm at the three stations. From the inversion results shown in Fig. 11(a), the VDR color ratio increases in turn at stations 1, 2, and 3, which probably indicates that the seawater particle components at the three stations are different. Because of the short wavelength 355 nm is more polarization sensitive to the small size particles than 532 nm, the higher VDR color ratio indicates the higher proportion of the small size particles. In addition, the K_lidar color ratio (Fig. 11(b)) shows good consistency with the CDOM (Fig. 11(c)) and Chl_a concentration. In the clean open ocean, the CDOM is mainly derived from the phytoplanktons, with extremely low concentration and the polarization-insensitive characteristic, which make it difficult to be observed by the single-wavelength of the VDR or the K_lidar. Since the CDOM has stronger absorption in the ultraviolet spectrum than that in the visible spectrum, which makes the 355 nm K_lidar larger than the 532 nm K_lidar, the K_lidar color ratio is always greater than 1 along the depth, and shows the similar trend as the CDOM profile. As shown in Fig. 11(c), between the depth of 10 m and 18 m, the CDOM concentration at station 3 is lower than that at stations 1 and 2, which corresponds to the minimum K_lidar color ratio. Below the depth of 18 m, the high CDOM concentration at station 3 (with the maximum value at about the depth of 25 m) leads to a large K_lidar color ratio. These results indicate that the dual-wavelength polarized ocean lidar is capability of obtaining multi-factor ocean parameters compared with single-wavelength ocean lidar.

Fig. 11. The comparison of the LOOP measurements of VDR color ratio and K_lidar color ratio with in-situ measurements of CDOM and Chl_a concentration. The blue, red, yellow, and green lines refer to the VDR color ratio, K_lidar color ratio, CDOM, and Chl_a concentration, respectively. The solid, lines, and dotted lines refer to the measurements at stations 1, 2, and 3, respectively.

Download Full Size | PDF

6. Conclusion

As a promising technology for ocean color remote sensing, the ocean lidar faces many challenges. The vertical profile of seawater optical parameters and Chl_a concentration is essential and lack of effective measurements. The LOOP system during the western Pacific campaign detected the seawater biological and optical parameters down to the depth of 50 m, but still not adequate to reveal the full profile of stratified phytoplankton layers or even to the bottom of mixed layer. Moreover, the magnitude of the lidar backscattered signal to the depth of 50 m reaches over five orders (clean open ocean), requiring the detectors and digitizers to have sufficient dynamic range. The saturating maximum counting rate of the photon detector and the resolution of the digitizer restrict the maximum detectable seawater depth in the upper-ocean mixed layer, where the most active ocean physical and chemical processes exist. The complex environmental conditions and compositions in the seawater make it difficult for lidar to analyze the single factor from the backscattered signal profile. The inversion results may be an integrated performance of various complex factors.

However, with limited but important information, the specially designed LOOP system provides new experience in the subsurface ocean exploration. LOOP is a dual-wavelength, variable-FOV, and polarized shipborne ocean lidar, which is the first kind of ocean lidar capable of obtaining the seawater optical parameters, the lidar VDR and the color ratio profiles of various compositions in the subsurface ocean. The high-power pulsed laser and photon-counting detector effectively increase the lidar detectable seawater depth in the clean open ocean. By controlling the peak pulse intensity of the backscattered signal, and calibrating the after-pulse effect of the PMT, the LOOP system ensures the linear response of the PMT for providing the seawater backscattered signal profiles. The retrieval of the FOV dependent IOP and AOP is a challenge for shipborne ocean lidar. Based on the LOOP design of variable FOV, the K_lidar with narrow FOV and wide FOV are compared with the in-situ measurement of seawater optical properties. The consistency of the results indicates that the ocean lidar can obtain the seawater optical properties in a certain depth, verifying the reliability of the LOOP system. Similarly, the VDR with narrow FOV and wide FOV combined with K_lidar explores the multiple scattering effect of ocean lidar detection in deep seawater. The good correlation between the lidar 532 nm VDR and the in-situ measurements of Chl_a concentration profiles at four stations indicates the possibility of distinguishing the type and concentration of particulate matter in seawater. Besides, the K_lidar and VDR at dual-wavelength of 355 nm and 532 nm are compared to explore the laser propagation characteristics in seawater, and the K_lidar color ratio is consistent with the in-situ measurements of CDOM profiles. For the further study, the combination of the K_lidar, the VDR, the K_lidar color ratio, and the VDR color ratio based on dual-wavelength will be utilized to identify the concentration and classification of various ocean compositions, which is one of the most challenging scientific issues in traditional ocean remote sensing. In addition, the time changing of the VDR and the color ratio vertical profiles, which will be a promising technology for continuous cross-section survey of ocean mixed layer ecosystem, provides the possibility for observing the spatiotemporal characteristics of various compositions in seawater based on the dual-wavelength ocean lidar. The fluorescence and Raman signals measured by the LOOP system in this campaign, which need more rigorous analysis and verification, are not introduced in this study. In future research, with more measurements of the LOOP system organized a new voyage in the South China Sea, we expect more detailed analysis mining on the detection capability of dual-wavelength ocean lidar, such as inelastic scattering detection for phytoplankton fluorescence, organic or inorganic compositions separation.

In summary, a dual-wavelength, variable-FOV, and polarized ocean lidar shows its capability of a deeper penetration depth and a day-and-night detection which leads to a layered characterization of the bio-optical properties in the subsurface ocean.

Funding

National Key Research and Development Program of China (2016YFC1400904); Pilot National Laboratory for Marine Science and Technology (2015ASTP-OS15, 2018ASKJ01); Key Technology Research and Development Program of Shandong Province (International Science and Technology Cooperation) (2019GHZ023); National Natural Science Foundation of China (61975191); China Postdoctoral Science Foundation (2021TQ0313).

Acknowledgments

We thank our colleagues for their kind support of this work, including the crews of R/V Haida for providing excellent assistance during the field campaign; Shuguo Chen from OUC for the excellent in-situ optical parameter measurements for the validation; Tong Cui, Xiangcheng Chen from OUC for their hard work received in the field investigations; Zhongping Lee from University of Massachusetts Boston for insightful discussion on IOPs and AOPs in clean water; Zhiyu Zhang for the advices on photomultiplier performance and editorial revision of the manuscript. We also acknowledge the use of data from NASA's Fire Information for Resource Management System (FIRMS) (https://earthdata.nasa.gov/firms), part of NASA's Earth Observing System Data and Information System (EOSDIS).

Disclosures

The authors declare no conflicts of interest.

Data availability

Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.

References

1. J. H. Churnside, “Review of profiling oceanographic lidar,” Opt. Eng. 53(5), 051405 (2013). [CrossRef]

2. C. Jamet, A. Ibrahim, Z. Ahmad, F. Angelini, M. Babin, M. J. Behrenfeld, E. Boss, B. Cairns, J. Churnside, J. Chowdhary, A. B. Davis, D. Dionisi, L. Duforêt-Gaurier, B. Franz, R. Frouin, M. Gao, D. Gray, O. Hasekamp, X. He, C. Hostetler, O. V. Kalashnikova, K. Knobelspiesse, L. Lacour, H. Loisel, V. Martins, E. Rehm, L. Remer, I. Sanhaj, K. Stamnes, S. Stamnes, S. Victori, J. Werdell, and P.-W. Zhai, “Going Beyond Standard Ocean Color Observations: Lidar and Polarimetry,” Front. Mar. Sci. 6, 251 (2019). [CrossRef]

3. A. P. Vasilkov, Y. A. Goldin, B. A. Gureev, F. E. Hoge, R. N. Swift, and C. W. Wright, “Airborne polarized lidar detection of scattering layers in the ocean,” Appl. Opt. 40(24), 4353–4364 (2001). [CrossRef]

4. J. H. Churnside, “Polarization effects on oceanographic lidar,” Opt. Express 16(2), 1196–1207 (2008). [CrossRef]

5. J. H. Churnside, J. M. Sullivan, and M. S. Twardowski, “Lidar extinction-to-backscatter ratio of the ocean,” Opt. Express 22(15), 18698–18706 (2014). [CrossRef]

6. J. H. Churnside and R. D. Marchbanks, “Inversion of oceanographic profiling lidars by a perturbation to a linear regression,” Appl. Opt. 56(18), 5228–5233 (2017). [CrossRef]

7. B. L. Collister, R. C. Zimmerman, C. I. Sukenik, V. J. Hill, and W. M. Balch, “Remote sensing of optical characteristics and particle distributions of the upper ocean using shipboard lidar,” Remote Sens. 215, 85–96 (2018). [CrossRef]

8. B. L. Collister, R. C. Zimmerman, V. J. Hill, C. I. Sukenik, and W. M. Balch, “Polarized lidar and ocean particles: insights from a mesoscale coccolithophore bloom,” Appl. Opt. 59(15), 4650–4662 (2020). [CrossRef]

9. M. J. Behrenfeld, Y. Hu, R. T. O’Malley, E. S. Boss, C. A. Hostetler, D. A. Siegel, J. L. Sarmiento, J. Schulien, J. W. Hair, X. Lu, S. Rodier, and A. J. Scarino, “Annual boom–bust cycles of polar phytoplankton biomass revealed by space-based lidar,” Nat. Geosci. 10(2), 118–122 (2017). [CrossRef]

10. M. J. Behrenfeld, P. Gaube, A. Della Penna, R. T. O’Malley, W. J. Burt, Y. Hu, P. S. Bontempi, D. K. Steinberg, E. S. Boss, D. A. Siegel, C. A. Hostetler, P. D. Tortell, and S. C. Doney, “Global satellite-observed daily vertical migrations of ocean animals,” Nature 576(7786), 257–261 (2019). [CrossRef]

11. P. Chen, C. Jamet, Z. Zhang, Y. He, Z. Mao, D. Pan, T. Wang, D. Liu, and D. Yuan, “Vertical distribution of subsurface phytoplankton layer in South China Sea using airborne lidar,” Remote Sens. 263, 112567 (2021). [CrossRef]

12. G. Chen, J. Tang, C. Zhao, S. Wu, F. Yu, C. Ma, Y. Xu, W. Chen, Y. Zhang, J. Liu, and L. Wu, “Concept Design of the “Guanlan” Science Mission: China’s Novel Contribution to Space Oceanography,” Front. Mar. Sci. 6, 194 (2019). [CrossRef]

13. D. M. Winker, W. H. Hunt, and C. A. Hostetler, “Status and performance of the CALIOP lidar,” in Laser Radar Techniques for Atmospheric Sensing, U. N. Singh, ed. (SPIE, 2004), 5575, pp. 8–15.

14. S. Wu, X. Song, B. Liu, G. Dai, J. Liu, K. Zhang, S. Qin, D. Hua, F. Gao, and L. Liu, “Mobile multi-wavelength polarization Raman lidar for water vapor, cloud and aerosol measurement,” Opt. Express 23(26), 33870–33892 (2015). [CrossRef]

15. A. Ansmann, M. Riebesell, and C. Weitkamp, “Measurement of atmospheric aerosol extinction profiles with a Raman lidar,” Opt. Lett. 15(13), 746–748 (1990). [CrossRef]

16. G. Dai, S. Wu, and X. Song, “Depolarization Ratio Profiles Calibration and Observations of Aerosol and Cloud in the Tibetan Plateau Based on Polarization Raman Lidar,” Remote Sens. 10(3), 378 (2018). [CrossRef]

17. Karyampudi, V. Mohan, Palm, and P. Stephen, “Validation of the Saharan Dust plume conceptual model using lidar, meteosat and ECMWF data,” Bulletin of the American Meteorological Society (1999).

18. C. Pérez, S. Nickovic, J. M. Baldasano, M. Sicard, F. Rocadenbosch, and V. E. Cachorro, “A long Saharan dust event over the western Mediterranean: Lidar, Sun photometer observations, and regional dust modeling,” J. Geophys. Res.: Atmos. 111(D15), D15214 (2006). [CrossRef]

19. R. E. Mamouri and A. Ansmann, “Fine and coarse dust separation with polarization lidar,” Atmos. Meas. Tech. 7(11), 3717–3735 (2014). [CrossRef]

20. X. Lu, Y. Hu, M. Vaughan, S. Rodier, C. Trepte, P. Lucker, and A. Omar, “New attenuated backscatter profile by removing the CALIOP receiver’s transient response,” J. Quant. Spectrosc. Radiat. Transfer 255, 107244 (2020). [CrossRef]

21. S. Chen, C. Xue, T. Zhang, L. Hu, G. Chen, and J. Tang, “Analysis of the Optimal Wavelength for Oceanographic Lidar at the Global Scale Based on the Inherent Optical Properties of Water,” Remote Sens. 11(22), 2705 (2019). [CrossRef]

22. L. Bingyi, L. Ruiqi, Y. Qian, and K. Xiaojuan, “Estimation of global detection depth of spaceborne oceanographic lidar in blue-green spectral region,” Infrared Laser Eng. 48(1), 106006 (2019). [CrossRef]

23. J. Alvarez, M. A. Vaughan, C. A. Hostetler, W. Hunt, and D. M. Winker, “Calibration technique for polarization-sensitive lidars,” J Atmos Ocean Technol. 23(5), 683–699 (2006). [CrossRef]

24. J. H. Churnside, “Review of profiling oceanographic lidar (erratum),” Opt. Eng. 56(7), 079802 (2017). [CrossRef]

25. H. R. Gordon, “Interpretation of airborne oceanic lidar: effects of multiple scattering,” Appl. Opt. 21(16), 2996–3001 (1982). [CrossRef]

26. D. M. Phillips and B. W. Koerber, “A Theoretical Study of an Airborne Laser Technique for determining Sea Water Turbidity,” Aust. J. Phys. 37(1), 75–90 (1984). [CrossRef]

27. R. E. Walker and J. W. McLean, “Lidar equations for turbid media with pulse stretching,” Appl. Opt. 38(12), 2384–2397 (1999). [CrossRef]

28. Z. Lee, J. Wei, K. Voss, M. Lewis, A. Bricaud, and Y. Huot, “Hyperspectral absorption coefficient of “pure” seawater in the range of 350–550 nm inverted from remote sensing reflectance,” Appl. Opt. 54(3), 546–558 (2015). [CrossRef]

29. J. D. Mason, M. T. Cone, and E. S. Fry, “Ultraviolet (250–550 nm) absorption spectrum of pure water,” Appl. Opt. 55(25), 7163–7172 (2016). [CrossRef]

30. A. Bricaud, A. Morel, M. Babin, K. Allali, and H. Claustre, “Variations of light absorption by suspended particles with chlorophyll a concentration in oceanic (case 1) waters: Analysis and implications for bio-optical models,” J. Geophys. Res.: Oceans 103(C13), 31033–31044 (1998). [CrossRef]

31. A. Morel, “Are the empirical relationships describing the bio-optical properties of case 1 waters consistent and internally compatible?” J. Geophys. Res.: Oceans 114(C1), C01016 (2009). [CrossRef]

32. A. Morel and B. Gentili, “The dissolved yellow substance and the shades of blue in the Mediterranean Sea,” Biogeosciences 6(11), 2625–2636 (2009). [CrossRef]

33. A. Morel and L. Prieur, “Analysis of variations in ocean color: Ocean color analysis,” Limnol. Oceanogr. 22(4), 709–722 (1977). [CrossRef]

34. H. Loisel and A. Morel, “Light scattering and chlorophyll concentration in case 1 waters: A reexamination,” Limnol. Oceanogr. 43(5), 847–858 (1998). [CrossRef]

35. V. Freudenthaler, M. Esselborn, M. Wiegner, B. Heese, M. Tesche, A. Ansmann, D. MüLLER, D. Althausen, M. Wirth, A. Fix, G. Ehret, P. Knippertz, C. Toledano, J. Gasteiger, M. Garhammer, and M. Seefeldner, “Depolarization ratio profiling at several wavelengths in pure Saharan dust during SAMUM 2006,” Tellus B 61(1), 165–179 (2009). [CrossRef]

36. K. Li, Y. He, J. Ma, Z. Jiang, C. Hou, W. Chen, X. Zhu, P. Chen, J. Tang, S. Wu, F. Liu, Y. Luo, Y. Zhang, and Y. Chen, “A Dual-Wavelength Ocean Lidar for Vertical Profiling of Oceanic Backscatter and Attenuation,” Remote Sens. 12(17), 2844 (2020). [CrossRef]

37. A. Morel and S. Maritorena, “Bio-optical properties of oceanic waters: A reappraisal,” J. Geophys. Res.: Oceans 106(C4), 7163–7180 (2001). [CrossRef]

38. J. M. Sullivan and M. S. Twardowski, “Angular shape of the oceanic particulate volume scattering function in the backward direction,” Appl. Opt. 48(35), 6811–6819 (2009). [CrossRef]

39. J. H. Churnside, “Airborne lidar estimates of photosynthesis profiles,” in 2016 IEEE International Geoscience and Remote Sensing Symposium (IGARSS) (2016), pp. 3777–3780.

40. D. Liu, P. Xu, Y. Zhou, W. Chen, B. Han, X. Zhu, Y. He, Z. Mao, C. Le, P. Chen, H. Che, Z. Liu, Q. Liu, Q. Song, and S. Chen, “Lidar Remote Sensing of Seawater Optical Properties: Experiment and Monte Carlo Simulation,” IEEE Trans. Geosci. Remote Sensing 57(11), 9489–9498 (2019). [CrossRef]

41. Q. Liu, X. Cui, W. Chen, C. Liu, J. Bai, Y. Zhang, Y. Zhou, Z. Liu, P. Xu, H. Che, and D. Liu, “A semianalytic Monte Carlo radiative transfer model for polarized oceanic lidar: Experiment-based comparisons and multiple scattering effects analyses,” J. Quant. Spectrosc. Radiat. Transfer 237, 106638 (2019). [CrossRef]

Subsystem	Parameter	Description
Transmitter	Pulse energy	200 mJ @ 532 nm
	Pulse energy	120 mJ @ 355 nm
	Repetition rate	20 Hz
	Pulse duration	≤ 5 ns
	Beam diameter	≈ 6.5 mm
	Beam divergence	≤ 0.5 mrad
	Polarization	Vertical @ 532 nm
	Polarization	Horizontal @ 355 nm
	Polarization ratio	> 100: 1

Subsystem	Parameter	Description
Telescope	Aperture diameter (D)	60 mm
	FOV	16∼60 mrad
	Range to full overlap (O_z)	21.90 m @ 16 mard
9.70 m @ 60 mard
Spectroscopy	Filter-bandwidth (FWHM)	2 nm @355, 532, 480 nm,10 nm @ 680 nm
Spectroscopy	Channels	355 nm parallel & orthogonal532 nm parallel & orthogonal404 nm & 680 nm
Polarizing beamsplitters	Transmission efficiency (T_p)	> 95%
	Reflection efficiency (R_s)	> 99.5%
	AR coating reflectance (R_abs)	< 0.25%
	Extinction ratio (T_p: T_s)	> 10⁶: 1

Subsystem	Parameter	Description
Detector and recorder	Neutral-density filters (NDF)	OD 2 @ 355 nm
		OD 5 @ 532 nm
		OD 4 @ 404 nm
		OD 3 @ 680 nm
	Photomultiplier tube	Hamamatsu H10721P-210
	AD resolution	14 bits
	Sampling frequency	PC @ 1 GHz

Instrument	Parameter	Description
BB9-FL3 (WET Labs Inc.)	Chl_a	Chlorophyll a concentration
BB9 (WET Labs Inc.)	b_bp	Total particulate backscattering coefficient
BB9 (WET Labs Inc.)	CDOM	Chromophoric dissolved organic matter
ac-s (WET Labs Inc.)	a_nw	Total non-water absorption coefficient
RAMSES (TriOS)	K_d	Diffuse attenuation coefficient

Stations	Location	Channels	Time
Station.1	127°E, 21°N	355 nm & 532 nm @ 60 mrad	20:30–23:30 / 2019.10.23
Station.2	127°E, 20°N	355 nm & 532 nm @ 60 mrad	15:10–18:10 / 2019.10.25
Station.2	127°E, 20°N	355 nm & 532 nm @ 16 mrad	18:25–23:10 / 2019.10.25
Station.3	127°E, 19°N	355 nm & 532 nm @ 60 mrad	15:30–18:30 / 2019.10.26
Station.4	130°E, 12°N	532 nm @ 60 mrad	15:05–18:30 / 2019.11.15
Station.4	130°E, 12°N	532 nm @ 16 mrad	18:35–23:35 / 2019.11.15

Shipborne variable-FOV, dual-wavelength, polarized ocean lidar: design and measurements in the Western Pacific

Abstract

1. Introduction