Abstract:

The droplet size distribution (DSD) formed by gas-saturated oil jets is one of the most important characteristics of the flow to understand and model the fate of uncontrolled deep-sea oil spills. The shape of the DSD, generally modeled as a theoretical lognormal, Rosin-Rammler or non-fundamental distribution function, defines the size and the mass volume range of the droplets. Yet, the fundamental DSD shape has received much less attention than the volume median size (d50) and range of the DSD during ten years of research following the Deepwater Horizon (DWH) blowout. To better understand the importance of the distribution function of the droplet size we compare the oil rising time, surface oil mass, and sedimented and beached masses for different DSDs derived from the DWH literature in idealized and applied conditions, while keeping d50 constant. We highlight substantial differences, showing that the probability distribution function of the DSD for far-field modeling is, regardless of the d50, consequential for oil spill response. This research was made possible by a grant from the Gulf of Mexico Research Initiative (GoMRI) C-IMAGE III and RECOVER2 Consortia. This dataset supports the publication: Faillettaz, R., Paris, C. B., Vaz, A. C., Perlin, N., Aman, Z. M., Schlüter, M., & Murawski, S. A. (2021). The choice of droplet size probability distribution function for oil spill modeling is not trivial. Marine Pollution Bulletin, 163, 111920. doi:10.1016/j.marpolbul.2020.111920

Suggested Citation:

Faillettaz, Robin, Claire B. Paris, Ana C. Vaz, Natalie Perlin, Zachary M. Aman, Michael Schlüter, and Steve Murawski. 2020. Dataset for: The choice of droplet size probability distribution function for oil spill modeling is not trivial. Distributed by: GRIIDC, Harte Research Institute, Texas A&M University–Corpus Christi. doi:10.7266/EZYAADMS

Publications:

Faillettaz, R., Paris, C. B., Vaz, A. C., Perlin, N., Aman, Z. M., Schlüter, M., & Murawski, S. A. (2021). The choice of droplet size probability distribution function for oil spill modeling is not trivial. Marine Pollution Bulletin, 163, 111920. doi:10.1016/j.marpolbul.2020.111920

Purpose:

Demonstration of the importance of oil droplet size distribution when modeling the dispersal of oil, in particular following an uncontrolled deep-sea blowout such as the Deepwater Horizon oil spill.

Data Parameters and Units:

For every two simulated cases of oil droplet size distribution range (case 1: 1-2400 µm and case 2: 1-500 µm), three probability distribution functions are provided. For each six, the mass of oil in the surface layer (0-1 m) and that of oil that landed+sedimented are provided for each 0.1°x0.1° bin in the Gulf of Mexico. Parameters for the oil mass files are time (days after the initial blowout, [days]), lat (latitude at the grid cell center, [decimal degrees N]), lon (longitude at the grid cell center, [decimal degrees E]), zbound (vertical layer bounds, [m]), depth (bathymetry, [m]), oil_conc (oil concentration, [ppb]). Parameters for the landed oil mass files are time (since release, [seconds]), depth [m], latitude [decimal degrees], longitude [decimal degrees], Landed oil mass [kg]. Parameters in the DSD files are droplet size (diameter, [µm]), frequency probability [unitless].

Methods:

The oil-CMS was used to simulate the different cases of oil droplet sizes. The simulation began April 20, 2010, 0000 UTC, with particles tracked for 100 days. Oil particles were released at 28.736°N, 88.365°W, at a depth of 1222m, or 300m above the oil well. 3000 particles were released every 2 hours, for 87 days, equivalent to a total of 3,132,000 oil particles released during the simulation. The transport and evolution of the oil particles were tracked by the oil-CMS model during the 100 days of the simulation, recording each particle’s horizontal position, depth, diameter, and density into the model output every two hours. Post-processing of the model data to compute the total mass per 0.1°x0.1° bins over the entire duration of the oil spill was conducted with Matlab, and data analysis with R. See the associated paper (Faillettaz et al, 2021) for more details on the methods.

Individual Files: