Abstract:

Numerical simulations are performed to study the effect of oil plumes on upper-ocean light field. The simulations are done using a set of numerical models that are designed to capture the detailed transport phenomena of oil by turbulence and waves in the ocean mixed layer and the light transfer through this complex medium. Particularly, the dispersion of the oil plume is simulated using the large-eddy simulation method, the sea-surface wave is simulated using the high-order spectral method, and the sunlight field variation is simulated using a Monte Carlo radiative transfer model. The simulation results show that, when interacting with flows in the ocean mixed layer, oil plumes of different droplet sizes exhibit very different dilution patterns in terms of plume size, shape and oil concentration. With the same volumetric release rate, plumes of large oil droplets exhibit highly intermittent local preferential concentration in near-surface regions and cause strong but intermittent sunlight extinction over relatively small horizontal area; plumes of small oil droplets diffuse more smoothly over large horizontal and vertical extensions, resulting in much more significant overall extinction effect to the sunlight penetration in the ocean euphotic zone. This dataset is associated with the manuscript "Effect of oil plumes on upper-ocean radiative transfer - a numerical study" submitted to Ocean Modelling.

Suggested Citation:

Xiao, Shuolin; Yang, Di. 2019. Effect of oil plumes on upper-ocean radiative transfer – a numerical study. Distributed by: Gulf of Mexico Research Initiative Information and Data Cooperative (GRIIDC), Harte Research Institute, Texas A&M University–Corpus Christi. doi:10.7266/n7-w0hm-6f39

Publications:

Xiao, S., & Yang, D. (2020). Effect of oil plumes on upper-ocean radiative transfer — A numerical study. Ocean Modelling, 145, 101522. doi:10.1016/j.ocemod.2019.101522

Purpose:

This dataset was generated by numerical simulations for studying the effect of surface oil plumes on ocean light field variation.

Data Parameters and Units:

The dataset contains a collection of subsets of data files that are organized under 12 different sub-directories. The description of the dataset structure are as follow: 1) The folder “3D Overall Configuration” contains the 3D oil concentration field (in “Concentration 3D.dat”), 2D downward irradiance fields (in “Downward Irradiance x=394m.dat” and “Downward Irradiance z=-96m.dat”), and 3D surface waves (in “HOS Surface Wave.dat”) of the LC flow condition. This dataset is for illustration of the overall configuration of the physical problem. The coordinates x, y and z are in the unit of meter; the downward irradiance E is normalized by the mean surface value and has no unit; the oil mass concentration C is for the oil droplets of diameter d = 0.42 mm and has the unit of kg/m3. 2) The folder “HOS Wave” contains the 1D wave spectrum, instantaneous wave elevation, and the corresponding wave Stokes drift for the condition of the 10 m/s wind. The coordinates x, y and z are in the unit of meter; the wavenumber “k dimensional” has the unit of m−1; and the wave spectral density “SWK1D” has the unit of m4. 3) The folder “3D LES Illustration” contains the instantaneous vertical velocity w from the LC case. Three representative 2D planes are extracted from the same snapshot. The coordinates x, y and z are in the unit of meter; and the vertical velocity w is in the unit of m/s. 4) The folder “Broadband Wave Case” contains the data for the downward irradiance of the 500 nm light under broadband sea-surface waves. The horizontal size of the sea-surface wave field is 512m × 512m, and the instan- taneous wave surface elevation data is contained in the file “Broadband Wave Case Wave Elevation.dat”. In this simulation case, the inherent optical properties of the seawater are (a, b) = (0.0257, 0.0029) m−1 and dif- fuse attenuation coefficient is Kd = 0.0271 m−1 . The instantaneous downward irradiances at two water depths (i.e. z = −25 and −50 m) are included in the files “Broadband Wave Case Downward Irradiance z=-25m.dat” and “Broadband Wave Case Downward Irradiance z=-50m.dat”. In these three data files, the coordinates x, y and z are in the unit of meter, and the downward irradiance E is normalized by the mean surface value (thus has no unit). In addition, the vertical profile of the horizontally averaged downward irradiance is included in “Broadband Wave Case Horizontal Averaged Downward Irradiance.dat”, and a reference profile based on the exponential decay of exp(Kd z) is included in “Broadband Wave Case Reference Downward Irradiance.dat”.For these two vertical profiles, the downward irradiance is normalized by its horizontal mean at z = −1 m. 5) The folder “OML Oil 3D Illustration” contains the instantaneous vertical velocity w and surface oil concen- tration C for both the LC and CC cases. For each case, the data contains the distributions of w on three representative 2D planes at x = 0 m, y = 0 m and z = −6 m, as well as the mass concentration for the oil droplets of diameter d = 0.42 mm on the 2D plane at z = −3 m. The coordinates x, y and z are in the unit of meter; the vertical velocity w is in the unit of m/s; and the oil mass concentration C is in the unit of kg/m3. 6) The folder “2D surface Plume” contains the 2D distributions of the instantaneous vertical velocity w and the corresponding surface mass concentrations PCon1–PCon3 for the oil droplets of diameters d = 0.27, 0.42 and 0.70 mm, respectively. The coordinates x, y and z are in the unit of meter; the vertical velocity w is in the unit of m/s; and the oil mass concentrations PCon1–PCon3 are in the unit of kg/m3. 7) The folder “Ed 2D Instant” contains the 2D surface oil concentration C at z = 0 m and the corresponding downward irradiance E at z = −10 m for both the LC and CC cases for two different droplet diameters d = 0.27mm (Particle1) and d = 0.70mm (Particle3). The coordinates x, y and z are in the unit of meter; the surface oil mass concentrations C is in the unit of kg/m3; and the downward irradiance E is normalized by the mean surface value and has no unit. 8) The folder “Ed Mean Profiles LC” contains the vertical profiles of horizontal average statistics for cases under the LC flow condition, including the oil-contaminated area A (normalized by the total horizontal area), the oil mass concentration C (in kg/m3), the light beam attenuation coefficient c (in m−1), and the deficit of downward irradiance E deficit of the 450 nm light (normalized by the corresponding horizontally averaged downward irradiance at the sea surface). The results for the oil droplets of diameters d = 0.27, 0.42 and 0.70 mm are labeled as Particle1, Particle2 and Particle3, respectively. For comparison, the vertical profile of the light beam attenuation coefficient c (in m−1) of natural seawater is also included. The coordinate z is in the unit of meter. 9) The folder “Ed Mean Profiles CC” contains the vertical profiles of horizontal average statistics for cases under the CC flow condition, including the oil-contaminated area A (normalized by the total horizontal area), the oil mass concentration C (in kg/m3), the light beam attenuation coefficient c (in m−1), and the deficit of downward irradiance E deficit of the 450 nm light (normalized by the corresponding horizontally averaged downward irradiance at the sea surface). The results for the oil droplets of diameters d = 0.27, 0.42 and 0.70 mm are labeled as Particle1, Particle2 and Particle3, respectively. For comparison, the vertical profile of the light beam attenuation coefficient c (in m−1) of natural seawater is also included. The coordinate z is in the unit of meter. 10) The folder “Chlorophyll Contained Seawater Validation” contains the vertical profile of horizontally aver- aged downward irradiance E of the 500 nm light in natural seawater obtained from the Monte Carlo simulation based on the empirical chlorophyll concentration profile with a mean surface chlorophyll concentration of 1 mg/m3 (Wo´zniak et al., 2003). For comparison, the empirical vertical decay profile of E from Wo´zniak et al.(2003) is also included in the dataset. The coordinate z is in the unit of meter. The downward irradiances are normalized by the corresponding surface mean. 11) The folder “Scattering Validation” contains the dependence of the scattering phase function on the scattering angle calculated based on the Mie theory module in the current Monte Carlo radiative transfer simulation model. The results for three representative diffraction parameters of 5, 25, and 100 are included for illustration purpose. 12) The folder “MC Wave Effect” contains the vertical profile of horizontally averaged downward irradiance E of the 450 nm light in natural seawater for the LC and CC flow conditions without oil plumes. For comparison, the vertical profile of E in natural seawater without flat surface are also included. The coordinate z is in the unit of meter. The downward irradiances are normalized by the corresponding surface mean.

Methods:

Summary of simulation parameters: We consider the oil dispersion in the ocean mixed layer (OML) and the resulted light field variation under the influences of shear- induced turbulence, Langmuir circulations, sea surface waves, and thermal convections. We consider two different sea-surface forcing conditions corresponding to shear-dominant and convection-dominant conditions. In the shear- dominant case, a constant wind shear stress τw = 0.168 N m−2 is applied in the x-direction, which corresponds to a wind speed U10 = 10 m/s (measured at 10 m height) and a friction velocity u∗ = 1.28 cm/s in water based on empirical parameterization (Donelan, 1982). A weak heat flux of Q = −15.5 Wm−2 (out of the ocean) is imposed at the surface to help spin-up the flow. Under this condition, the OML flow is dominated by the three-dimensional turbulence generated by the wind-induced shear and the coherent Langmuir circulation cells generated by wave–turbulence interaction (the wave condition is discussed further below). In the rest of this paper, we refer to this flow condition as LC. In the convection-dominant case we apply a weaker wind stress τw = 0.036 N m−2 (corresponding to wind speed U10 = 5 m/s and friction velocity u∗ = 0.59 cm/s in water) and a stronger surface heat flux of Q = −207.0 Wm−2. Under this condition, the convective cells generated by the thermal convection are the dominant flow structures, and hereinafter we refer to this condition as CC. For both flow conditions, we set the Coriolis frequency to be fc = 7 × 10−5 s−1, corresponding to a latitude of 28.7◦ N. For both flow conditions, we assume the sea-surface wave field is in equilibrium with the wind forcing. We adopt a widely used empirical broadband wave spectrum parameterization (Donelan and Pierson, 1987). The wavelengths at the spectrum peak are λp = 92.2 m for LC and 23.1 m for CC, and the corresponding wave periods are Tp = 7.69 s and 3.85 s, respectively. Underneath the sea surface, we use a computational domain of 922 m long and wide and 100 m deep to simulate the oil dispersion in OML using LES with 384 × 384 × 257 computational grid points. The flow field is well mixed in the top half of the domain corresponding to an OML depth zi = 50 m, and stably stratified further below with a temperature gradient dΘ/dz = 0.01 K m−1. For the broadband wave conditions considered in this study, the corresponding turbulent Langmuir number is about the same for both the LC and CC conditions, Lat = u∗ /us,0 = 0.3, where u∗ is friction velocity caused by wind shear and us,0 is the magnitude of the wave-induced Stokes drift velocity at the mean water surface level (McWilliams et al., 1997). We use the reference seawater density ρ0 = 1031.0 kg/m3 and viscosity µ f =1.08×10−3 kg/(m s), and the oil density ρd = 859.9 kg/m3. For each flow condition, we consider three different cases differentiated by the droplet diameter, i.e. d = 0.27, 0.42 and 0.70 mm. For each case, a monodisperse oil plume with the same droplet size is released from a localized source at the 75 m depth with a low mass release rate of Qs = 10 kg/s. In the Monte Carlo radiative transfer simulation, the inherent optical properties (IOPs) induced by naturally existing substances in the seawater, such as water molecules, suspended particulate matter, and colored dissolved organic matter, are prescribed based on empirical parameterizations (Mobley, 1994; Wo´zniak and Dera, 2007; Kirk, 2011). The effects of oil droplets on the IOPs are modeled based on the simulated oil concentration using Mie theory (Mie, 1908; Bohren and Huffman, 2008) with the complex refractive index of spherical oil droplet set to be nd = 1.494 + 0.0089 i (Otremba and Piskozub, 2004). For the simulation cases with oil plumes, we simulated the radiative transfer of the 450 nm wavelength light, which is in the spectrum range for oceanic photosynthesis. For the validation cases without oil plumes, we simulated the radiative transfer of the 500 nm wavelength light in order to compare our numerical simulation results with the data available from the literature.

Provenance and Historical References:

Bohren, C. F., & Huffman, D. R. (1998). Absorption and Scattering of Light by Small Particles. doi:10.1002/9783527618156 Donelan, M. A. (1982). The dependence of the aerodynamic drag coefficient on wave parameters. In Proceedings of the First International Conference on Meteorology and Air-Sea Interaction of the Coastal Zone (pp. 381-387). American Meteorological Society, Boston, MA. Donelan, M. A., & Pierson, W. J. (1987). Radar scattering and equilibrium ranges in wind-generated waves with application to scatterometry. Journal of Geophysical Research, 92(C5), 4971. doi:10.1029/jc092ic05p04971 Kirk, J. T. (2011). Light and photosynthesis in aquatic ecosystems, 3rd Edition. Cambridge university press. McWilliams, J. C., Sullivan, P. P., & Moeng, C. H. (1997). Langmuir turbulence in the ocean. Journal of Fluid Mechanics, 334, 1–30. doi:10.1017/s0022112096004375 Mie, G. (1908). Beiträge zur Optik trüber Medien, speziell kolloidaler Metallösungen. Annalen Der Physik, 330(3), 377–445. doi:10.1002/andp.19083300302 Mobley, C.D. (1994). Light and Water: Radiative Transfer in Natural Waters, Academic Press. Otremba, Z., & Piskozub, J. (2004). Phase functions of oil-in-water emulsions. Optica Appl, 34(1), 93-99. Wozniak, B., & Dera, J. (2007). Light absorption in sea water (Vol. 33). New York, NY, USA: Springer. Wozniak, B., Dera, J., Ficek, D., Majchrowski, R., Ostrowska, M., & Kaczmarek, S. (2003). Modelling light and photosynthesis in the marine environment. Oceanologia, 45(2). Xiao, S., & Yang, D. (submitted to Ocean Modelling). Effect of oil plumes on upper-ocean radiative transfer - a numerical study.

Individual Files: