Visual Appearance of Oil on the Sea

Home , Bonn Agreement (1969), Color of water

Journal of Marine Science and Engineering

Review Visual Appearance of Oil on the Sea

Merv Fingas

Spill Science, Edmonton, AB T6W 1J6, Canada; ﬁ[email protected]

Abstract: The visual appearance of oil spills at sea is often used as an indicator of spilled oil properties, state and slick thickness. These appearances and the oil properties that are associated with them are reviewed in this paper. The appearance of oil spills is an estimator of thickness of thin oil slicks, thinner than a rainbow sheen (<3 µm). Rainbow sheens have a strong physical explanation. Thicker oil slicks (e.g., >3 µm) are not correlated with a given oil appearance. At one time, the appearance of surface discharges from ships was thought to be correlated with discharge rate and vessel speed; however, this approach is now known to be incorrect. Oil on the sea can sometimes form water-in-oil emulsions, dependent on the properties of the oil, and these are often reddish in color. These can be detected visually, providing useful information on the state of the oil. Oil-in-water emulsions can be seen as a coffee-colored cloud below the water surface. Other information gleaned from the oil appearance includes coverage and distribution on the surface.

Keywords: oil spill appearance; oil color; oil slick thickness; Bonn code; ASTM code

1. Introduction When an oil spill occurs, there is a need for extensive information including the location of the oil, its distribution, its trajectory and its thickness [1–3]. Thickness is of particular interest because many countermeasures are dependent on oil being a minimum thickness in order to be effective. This is true for in-situ burning and chemical oil spill dispersion, and, to a certain degree, for oil removal by skimmers. Citation: Fingas, M. Visual Observations in the visible spectrum, be they by human eyesight or through the use of Appearance of Oil on the Sea. J. Mar. cameras or spectrometers, have been used over the past decades to gain information about Sci. Eng. 2021 9 , , 97. https:// spilled oil [1,3]. The suggested uses of such observations have been used to estimate the doi.org/10.3390/jmse9010097 state of the oil, the position and distribution, its condition and thickness. The primary tool for visual observation of oil is to use its color or appearance. Most crude and fuel oils are Received: 3 December 2020 black; however, when very thin (<3 µm), oil has other appearances based on its interaction Accepted: 14 January 2021 Published: 18 January 2021 with light. These light interactions can change the color or appearance of the oil. Efforts have been made to correlate these colors and appearance with the thickness of oil. This

Publisher’s Note: MDPI stays neutral correlation of color and thickness will be reviewed in detail below. Further, it is noted that with regard to jurisdictional claims in there are variances to the appearance of oil caused by factors such as viewing angle, wind published maps and institutional affil- effects, solar illumination and water conditions. iations. Other uses of visual observations are suggested. These include observation of the formation of water-in-oil emulsions [4], indication of oil-in-water emulsions [5], measurement of subsea discharge rates [2] and the simple measurement of the oil geometry on the sea [6]. These observation uses will be reviewed in this paper. In the past, visual observations were standardized in various codes to provide users Copyright: © 2021 by the author. Licensee MDPI, Basel, Switzerland. with a standard way of interpreting appearances of oil on the sea [1]. These did not enter This article is an open access article the broad scientific literature; rather, they became industry and government practices. This distributed under the terms and paper will examine the appearance of the oil on the sea and provide insight into those conditions of the Creative Commons appearances that provide correct information about the state of the oil. Attribution (CC BY) license (https:// The appearance of oil on the sea is subject to several variances [1]. The most important creativecommons.org/licenses/by/ variable is that of oil type. Crude oils are typically brown to black when on the sea in 4.0/). thicker slicks (>3 µm) [1]. Fuel oils are black. The native oil color can have an influence on

J. Mar. Sci. Eng. 2021, 9, 97. https://doi.org/10.3390/jmse9010097 https://www.mdpi.com/journal/jmse J. Mar. Sci. Eng. 2021, 9, 97 2 of 13

the final appearance. Secondly, the viewing angle has an influence on appearance when viewing thin slicks [1]. In daylight, with the broad illumination that occurs, the viewing angle is not as important [1]. Atmospheric conditions have an effect on appearance [1]. For example, slick observations can be clearer when there is cloud cover. Atmospheric haze can affect the clarity of observations. Finally, sea conditions can have an influence [1]. Because oil slicks often cover vast areas, the effect of sea conditions on observations is minimized.

2. Thickness 2.1. Basic Considerations The knowledge of slick thickness has many uses in the oil spill trade, which include determining the size of a leak and optimizing the effectiveness of oil spill countermeasures [1]. Countermeasures such as chemical dispersion and in-situ burning are highly dependent on slick thickness. In the past, it was thought that there was thickness data in the color of oil slicks [1]. Over the years, experiments were conducted to show that there was only a relationship between appearance and thin slicks, that is, slicks less than 3 µm in thickness [1]. The start of this work was in 1930 [7]. Some recent work has been carried out; however, all of it has been carried out in close proximity to the slicks and not from airborne platforms. The appearance and color of slicks as related to slick thickness, are affected by several factors in addition to thickness [1]. Some of these are noted in Section1 above. In addition, there is the question of slick heterogeneity [1,8]. Oils may not form a consistent ﬂat slick over large extents on the water. Some experimental work has shown that some slicks form a “fried egg” type of proﬁle. Sometimes slicks do not cover the entire area of consideration. On other occasions, interfacial tension may herd spilled oil into multiple small slicks, instead of forming one consistently thick slick. In the absence of accurate slick thickness methods, visual judgement on slicks may be incorrect [9–11]. In summary, the relationship of oil thickness to appearance is subject to a number of variations [1,2]. The major question still needs to be answered, that is, is there a fundamental relationship between oil thickness and appearance? Furthermore, if there is, what ranges of oil thickness does it apply to and under what conditions?

2.2. Fundamental Slick Thickness–Color Relationships The appearance of an oil slick on water is dependent on several processes including reflection from the oil–air interface, the oil–water interface and various processes in the oil layer itself [10,11]. Some of these processes are illustrated in Figure1. Ambient sunlight can enter the oil and will be refracted by the oil as a result of the diffraction index of oil (typically about 1.5). Light then can be reflected back through the oil from the oil–water interface and back to the surface of the oil. For thin slicks, the reflectivity is high because oil absorption is low as a result of the low attenuation of the light by the thin slick. The light absorption by the oil becomes more significant as the oil becomes thicker, that is, above 3 µm [11]. Oils have varying absorbances of light, as will be shown later. An important process is the interference of light waves between that entering the oil and that being reflected from the oil/water interface. For the purposes of this paper, thin slicks are defined as those below 3 µm and thick slicks as those above 3 µm. The processes which influence the amount and nature of the light viewed by the observer are described below. The major influence on light returned to the observer is that of reflection. The layer of oil acts as a mirror and reflects light at the same angle as it was transmitted [12]. As some thin slicks are somewhat transparent to light, part of the light will be refracted into the oil and then again into the water column [13]. While light passes through the oil, some of the light is absorbed. J. Mar. Sci. Eng. 2021, 9, 97 3 of 13 J. Mar. Sci. Eng. 2021, 9, x FOR PEER REVIEW 3 of 13

FigureFigure 1. 1.Fundamental Fundamental optical optical phenomena phenomena involved involved in in the th oile oil coloration coloration of of thin thin slicks. slicks. This This diagram diagram is applicable only to the consideration of the appearance of thin slicks. is applicable only to the consideration of the appearance of thin slicks.

TheThe observerprocesses views which the influence oil somewhat the amount differently and nature if the of oil the is verylight thinviewed (<1 µbym) the or ob- if veryserver thick are (>10describedµm). Atbelow. various The thicknesses, major influence different on light physical returned phenomena to the observer predominate. is that Atof lessreflection. than 1 µ Them oil layer thickness, of oil acts the lightas a mirror returned and to reflects the observer light fromat the sunlight same angle is composed as it was oftransmitted reflection from[12]. As the some oil surface, thin slicks reflection are somewhat from the transparent water surface to andlight, some part upwellingof the light light.will be This refracted multiple into reflection the oil and makes then oil again appears into silvery.the water column [13]. While light passes throughAs the the oil oil, thickens some of past the about light 1isµ absorbed.m, the absorption of the oil begins to play a significant role inThe the observer appearance views of the the oil oil [ 9somewhat,11]. The absorption differently by if lightthe oil is is given very as thin [12 ]:(<1 µm) or if very thick (>10 µm). At various thicknesses, different physical phenomena predominate. −ax At less than 1 µm oil thickness, the lightI = re Ioturnede to the observer from sunlight is com-(1) posed of reflection from the oil surface, reflection from the water surface and some where:upwelling Io is light. the incident This multiple radiation, reflection I is the radiationmakes oil at appears a specified silvery. depth, a is the absorption coefficient,As the and oil thickens x is the depthpast about of the 1 µm, absorber. the abso Otrembarption of [14 the] measured oil begins theto play attenuation a signifi- ofcant two role oils, in Petrobaltic,the appearance a medium of the oil oil, [9,11]. and Romashkino,The absorption a darkerby light oil. is given These as absorption [12]: coefficients vary with wavelength, and the average for Petrobaltic was 5005 m−1, and − for Romashkino was 150,685 m 1. ExtendingI = theseIoe−ax averages over thicknesses of interest,(1) we arrive at Table1 using Equation (1). Absorption is shown as the absolute value of where: Io is the incident radiation, I is the radiation at a specified depth, a is the absorption attenuation, e.g., a value of 2 reduces the light by a factor of 2. coefficient, and x is the depth of the absorber. Otremba [14] measured the attenuation of two oils, Petrobaltic, a medium oil, and Romashkino, a darker oil. These absorption coef- Table 1. Light attenuation by absorption. ficients vary with wavelength, and the average for Petrobaltic was 5005 m−1, and for Ro- Scheme 0mashkino 0.3 was 0.6150,685 m 1.2−1. Extending 1.8 these averages 2.4 over 3 thicknesses 3.6 of 4.2interest, 4.5we arrive at TableAverage 1 using Absorption Equation over (1). Visible Absorption Spectrum is shown as the absolute value of attenua- Absorption Petrobaltiction, e.g., 1 a value 1 of 2 reduces 1 the light 1 by a factor 1 of 2. 1 1 1 1 Absorption Romashkino 1 1.1 1.2 1.3 1.4 1.6 1.7 1.9 2 Table 1. Light attenuation by absorption.

As canScheme be seen 0 from Table 0.31, absorption0.6 does1.2 not signiﬁcantly1.8 2.4 affect3 the3.6 attenuation 4.2 4.5 of 1 the Petrobaltic oil belowAverage 4.5 µm, butAbsorption the Romashkino over Visible has anSpectrum attention of 2; that is, about ⁄2 theAbsorption light is absorbed Petrobaltic at 4.5 µm. 1 1 1 1 1 1 1 1 1 AbsorptionAnother Romashkino phenomenon that1 takes1.1 place is1.2 the interference1.3 1.4 between1.6 light1.7 waves1.9 [132 ]. This interference results in the rainbow colors of thin slicks. The refractive index of a given wavelengthAs can resultsbe seen in from a difference Table 1, inabsorption the optical does path not length. significantly This difference affect the can attenuation be given as [13]: of the Petrobaltic oil below 4.5 µm, but the Romashkino1 has an attention of 2; that is, about ∆L = 2t (µ2 − sin2(I)) ⁄2 (2) ½ the light is absorbed at 4.5 µm.

J. Mar. Sci. Eng. 2021, 9, 97 4 of 13

where: ∆L is the difference in optical path length; t is the ﬁlm thickness; µ is the refractive index of the ﬁlm; I is the angle of light incidence. Hornstein [13] points out that if ∆L contains a whole number of wavelengths, then the maximum destructive interference will occur. If ∆L contains an odd number of half wavelengths, then the maximum constructive interference will occur. The maximum destructive interferences occur at:

λ = ∆L/x (3)

where λ is the wavelength under consideration, and x is a whole number as 1, 2, 3, 4 etc. The maximum constructive interferences occur at:

λ = 2∆L/x (4)

where x is a whole odd number as 1, 3, 5, 7 etc. This results in the interferences shown in Table2[13].

Table 2. Typical wave lengths resulting from interference [13].

Oil Thickness Constructive Wavelengths Destructive Wavelengths Resulting Color(s) µm nm nm 0.3 544 816, 408 First colors appear 0.6 652, 465 816, 544, 408 Bright Rainbow Colors 0.9 699, 544, 445 816, 612, 490, 408 1.2 725, 593, 502, 435 816, 652, 544, 466, 408 Dull Rainbow Colors 1.5 743, 528, 544, 480, 429 816, 680, 583, 510, 453, 408 More Dull Colors 1.8 753, 653, 616, 516, 466, 426 816, 699, 612, 544, 490, 445, 408

Table2 shows the colors that are the result of the light interference in the oil layer [ 13]. Furthermore, the dulling of the colors shows that with increasing numbers of constructive and destructive interferences, colors become increasing dull. The decrease in light, as judged by the color dullness, approximates double the number of interferences. This effect is additive to absorption. Calculating this effect for Petrobaltic, a medium oil, and Romashkino, a dark oil, shows that the oil colors appear earlier for the dark oil. This calculation is shown in Table3. The calculations begin at the bottom of Table3. The attenuation equivalent of the effect of number of constructive and destructive interferences is taken as a doubling of the attenuation with each doubling of the number of interferences, beginning with the bright colors at 0.9 µm and thereafter with 3 constructive interferences. The calculated absorptions for the two oils (similar to Table1) are added to the relative attenuations resulting from constructive and destructive interference. This gives rise to a relative attenuation number called the total attenuation in Table3. The end result is that the true oil color of the darker oil, Romashkino, appears at about 1.8 µm, whereas this true oil color appears at about 2.7 µm for the medium oil, Petrobaltic. Thus, at thicknesses greater than the rainbow colors, >1.5 µm, the true oil colors will appear sooner for a darker oil. Further, this calculation, as shown in Table3, shows that there are no possible colorations beyond about 3 µm for any oil. This is true even if the effect of constructive and destructive interferences for the oil is not considered. J. Mar. Sci. Eng. 2021, 9, 97 5 of 13

Table 3. Effect of absorption and interference.

Slick Thickness µm 0.3 0.6 0.9 1.2 1.5 1.8 2.1 2.4 2.7 3 3.3 3.6 3.9 4.2 4.5 10 20 Total Attenuation Petrobaltic 1 2 2 2.3 2.6 3 3.3 3.6 4 4.3 4.6 5 5.3 5.6 6 9.1 17.1 First colors appear Dull Rainbow Colors More Dull Colors Oil Color Oil Color Oil Color Oil Color Oil Color Oil Color Oil Color Oil Color Oil Color Bright Rainbow Colors Romashkino 2 3.1 4.1 5.2 6.3 7.3 8.4 9.4 10.5 11.6 12.6 13.7 14.8 15.9 17 27.5 66.4

First colors appear Dull Rainbow Colors Oil Color Oil Color Oil Color Oil Color Oil Color Oil Color Oil Color Oil Color Oil Color Oil Color Oil Color Oil Color Bright Rainbow Colors More Dull Colors Absorption 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1.1 1.1 Petrobaltic Absorption 1 1.1 1.1 1.2 1.3 1.3 1.4 1.4 1.5 1.6 1.6 1.7 1.8 1.9 2 4.5 20.4 Romashkino Effect of Interference 0 1 1 1.3 1.6 2 2.3 2.6 3 3.3 3.6 4 4.3 4.6 5 8 16 # constructive 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 23 46 # destructive 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 24 47 J. Mar. Sci. Eng. 2021, 9, x FOR PEER REVIEW 6 of 13

J. Mar. Sci. Eng. 2021, 9, 97 6 of 13

2.3.2.3. RainbowRainbow SlicksSlicks StudiesStudies have have shown shown that that rainbow rainbow sheens, sheens, aa particularparticular formform ofof slickslick appearanceappearance illus-illus- tratedtrated in in Figure Figure2 ,2, actually actually has has different different physical physical origins origins from from other other thickness–appearance thickness–appearance relationshipsrelationships [1[1,13,15].,13,15]. HornsteinHornstein [[13,15]13,15] studiedstudied thethe variousvarious slickslick phenomenaphenomena andand con-con- cludedcluded thatthat rainbowrainbow sheen was was the the result result of of constructive constructive and and destructive destructive interference interference be- betweentween light light reflected reﬂected from from the the oil oilsurface surface and andthat thatfrom from the water the water surface. surface. This is This a unique is a uniqueindicator indicator of oil thickness. of oil thickness.

FigureFigure 2. 2.A A photograph photograph of of a a rainbow rainbow sheen sheen taken taken from from an an aircraft aircraft at at an an altitude altitude of of about about 300 300 m. m. This colorThis ofcolor sheen of sheen has a stronghas a strong physical physical relation relation to 1 to to 1.5 1 µtom. 1.5 µm.

TablesTables ofof constructive and and destructive destructive wave wavelengthslengths can can be written be written as shown as shown in Table in Table2 [13,15].2[13, 15This]. taskThis results task results in a color in a chart color for chart visible for oil visible as: thickness oil as: thickness less than 0.15 less µm— than 0.15no colorµm—no apparent; color apparent; thickness thickness of 0.15 µm—warm of 0.15 µm—warm tone apparent; tone apparent; thickness thickness of 0.2 to 0.9 of 0.2µm— to 0.9varietyµm—variety of colors of (e.g., colors rainbow); (e.g., rainbow); and for andthickness for thickness greater greaterthan 0.9 than µm—colors 0.9 µm—colors of less pu- of lessrity. purity. This corresponds This corresponds to the toobserved the observed onset onsetnoted notedin experiments in experiments [2]. [2].

2.4.2.4. ColorColor CodesCodes ColorColor codescodes areare charts relating oil oil thickness thickness to to color. color. Many Many different different codes codes were were de- developed—largelyveloped—largely based based on on older older work, work, some some of which of which are arelisted listed in Table in Table 4 [16].4[ 16The]. Thefind- ﬁndingsings of the of theresearch research to date to date include include the thefollowing following points points [1,2]. [1 ,First,2]. First, the theappearance appearance of thin of thinslicks slicks is relatively is relatively consistent consistent across across the the ma manyny experiments experiments carried carried out out over many years.years. Second,Second, thethe mostmost reliablereliable thicknessthickness appearancesappearances areare forfor thethe thinthin sheensheen and and the the rainbow rainbow sheen.sheen. Third,Third, thethe utilityutility ofof thicknessesthicknesses derivedderived fromfrom oiloil colorcolor codescodes isis veryvery limitedlimited [[16].16]. µ TheThe onlyonly reliable reliable color color codes codes are are the the thin thin ones ones (<3 (<3 µm).m). MostMost ofof thethe oiloil is is located located in in the the µ thickthick portions portions of of the the slick slick (thickness (thickness > 3> 3m). µm). For For larger larger spills, spills, the appearancethe appearance codes codes are not are useful.not useful. Finally, Finally, there there is no is supporting no supporting evidence evidence for any forcolorations any colorations for the for thicker the thicker portions por- oftions the slick.of the slick.

J. Mar. Sci. Eng. 2021, 9, x FOR PEER REVIEW 7 of 13 J. Mar. Sci. Eng. 2021, 9, x FOR PEER REVIEW 7 of 13

J. Mar. Sci. Eng. 2021, 9, 97 7 of 13 Table 4. Summary of Appearance Color Codes. Visibility Thresholds (µm) Visibility Thresholds (µm) Table 4. Summary of AppearanceHeight ColorViewing Codes. Reference Year Oil Type Number Height Viewing Darkening Dull Dark Col- Reference Year Oil Type Number (m) Angle Minimum Silvery Rainbow Darkening Dull Dark Col- (m) Angle Minimum Silvery Rainbow Colors Colors ors Visibility Thresholds (µm) Colors Colors ors Viewing Congress [7] 1930 various incl.Type Bunker, fuel oil e Height >15 ship board oblique 0.1 Reference Year Congress [7] Oil 1930 various incl. Bunker,Number fuel oil e >15 shipAngle board oblique 0.1 Darkening Dull Allen et al. [17] 1969 crude—Santa Barbara e (m) multiple ns Minimum ns Silvery 0.05Rainbow to 0.18 0.23 to 0.75 1 to 2.5 2.5 toDark 5.5 Colors Allen et al. [17] 1969 crude—Santa Barbara e multiple ns ns 0.05 to 0.18 0.23 toColors 0.75 1 to 2.5 Colors 2.5 to 5.5 API [18] 1969 general l ns ns 0.04 0.08 0.15 to 0.3 1 2 Congress [7] 1930API various [18] incl. Bunker, 1969 fuel oil general e >15 lship board obliquens 0.1 ns 0.04 0.08 0.15 to 0.3 1 2 Hornstein [13] 1972 Arabian and Louisiana crudes e >20 1 to 2 various <0.15 up to 0.15 0.15 to 0.9 0.9 to 1.5 1.5 to 3 Allen et al. [17] 1969Hornstein crude—Santa [13] 1972 Barbara Arabian and eLouisiana multiple crudes e ns >20 ns 1 to 2 various 0.05 <0.15 to 0.18 up 0.23 to 0.15 to 0.75 0.15 to 1 0.9 to 2.5 0.9 to 2.5 1.5 to 5.5 1.5 to 3 Hornstein [15] 1973 various e & l ship & aerial various 0.038 0.076 0.15 to 0.31 1 2 API [18] 1969Hornstein [15] general 1973 various l e & l nsship ns & aerial various 0.04 0.038 0.08 0.076 0.15 to 0.3 0.15 to 0.31 1 1 2 2 Hornstein [13] 1972Parker Arabian et al. [19] and Louisiana 1979 crudes North Sea and e Arabian >20crudes e 1 to 2 2 ship various & aerial various <0.15 up 0.1 to 0.15 0.15 to 0.9 0.9 to 1.5 1.5 to 3 Parker et al. [19] 1979 North Sea and Arabian crudes e 2 ship & aerial various 0.1 Hornstein [15] 1973ITOPF [20] various 2005 egeneral & l ship l & aerial various aerial 0.038 ns 0.1 0.076 0.15 0.1 to 0.31 0.3 1 0.1 2 ITOPF [20] 2005 general l aerial ns 0.1 0.1 0.3 0.1 Parker et al. [19] 1979Schriel North [2] Sea and Arabian 1987 crudes general e 2 e ship& l & aerial various aerial various 0.1 0.05 0.1 0.15 0.3 1 2 ITOPF [20] 2005Schriel [2] general 1987 general l e & l aerial ns aerial various 0.1 0.05 0.1 0.1 0.3 0.15 0.1 0.3 1 2 Schriel [2] 1987 general e & l aerial various 0.1 0.3 1 5 15 Schriel [2] 1987Schriel [2] general 1987 e general & l e & l aerial various aerial various 0.05 0.10.1 0.15 0.3 0.3 1 1 5 2 15 Duckworth [21] 1993 various crudes e several ns ns 0.1 0.1 0.1 to 1 Schriel [2] 1987Duckworth [21] general 1993 various e & lcrudes e aerial several various ns ns 0.1 0.1 0.1 0.3 0.1 to 1 1 5 15 Brown et al. [22] 1995 crude—Norman Wells e 32 30 m nadir 0.094 Duckworth [21] 1993Brown et al. various[22] crudes 1995 crude—Norman e Wells several e ns 32 ns 30 m nadir 0.1 0.094 0.1 0.1 to 1 Brown et al. [22] 1995 crude—Norman Wells diesel e 32 e 30 m25 nadir30 m 0.094nadir 0.165 diesel e 25 30 m nadir 0.165 diesel lubricating e oil 25 e 30 m 16 nadir 30 m 0.165 nadir 0.077 lubricating oil e 16 30 m nadir 0.077 lubricating oilhydraulic e oil 16e 30 m13 nadir30 m 0.077nadir 0.159 hydraulic oil e 13 30 m nadir 0.159 hydraulic oil e 13 30 m nadir 0.159 Coast Guard [23] 1996 general—tar codes l 0.04 0.075 0.15 0.3 1 3 Coast Guard [23] 1996Coast Guard general—tar [23] 1996 codes general—tar l codes l 0.04 0.04 0.075 0.075 0.15 0.15 0.3 0.3 1 1 3 3 Bonn Agreement 50 to 200, 200 Bonn Agreement 2017 l 0.04 0.04 to 0.3 0.3 5 >5 to 50 b 5050 toto 200,200, 200200 b b bb Bonn Agreement [24] 2017[24] 2017 l l 0.04 0.04 0.04 to 0.3 0.04 to 0.30.3 0.3 5 5 >5 to >5 50 to 50 toto 20002000 b [24] to 2000 ASTM Standard [16] 2016 0.08 0.1 0.5 0.9 3 >3 ASTM Standard [16] 2016ASTM Standard [16] 2016 0.08 0.08 0.1 0.1 0.5 0.5 0.9 0.9 3 3 >3 >3 Average 0.09 0.1 0.6 0.9 2.7 8.5 Average Average 0.09 0.09 0.1 0.1 0.60.6 0.90.9 2.72.7 8.5 8.5 a b e = experiment; l = literature; ns = not specified; a dark is sometimes stated as “true oil color”, “black”, “brown” or “darker colors” or “metallic.”; b the Bonn agreement e = experiment; l a= literature; ns = not specified; dark is sometimes stated as “true oil color”, “black”, “brown”b or “darker colors” or “metallic.”; the Bonn agreement e = experiment; l = literature; nsdocument = not speciﬁed; has twodark additional is sometimes thicknesses, stated as based “true on oil color”,oil distribution: “black”, “brown” 50 to 200 or for “darker patchy, colors” discontinuous or “metallic.”; distributionthe Bonn and agreement >200 µ documentm for continuous has two additionalslicks. thicknesses, based on oil distribution: 50 todocument 200 for patchy, has two discontinuous additional distribution thicknesses, and based >200 onµm oil for distribution: continuous slicks. 50 to 200 for patchy, discontinuous distribution and >200 µm for continuous slicks.

J. Mar. Sci. Eng. 2021, 9, x FOR PEER REVIEW 8 of 13

J. Mar. Sci. Eng. 2021, 9, 97 8 of 13 One of the new color codes that has emerged has been that of the Bonn Agreement [24]. This is shown in Table 4. The part that is different from any other code is the thicker One of the new color codes that has emerged has been that of the Bonn Agreement indications.One of theThe new basis color for these codes thicker that has appearances emerged has is been some that experimental of the Bonn work; Agreement however, [24]. [24]. This is shown in Table 4. The part that is different from any other code is the thicker thisThis work is shown was never in Table reviewed4. The or part published that is different[25]. A second from issue any otheris the calling code is of the the thicker thicker indications. The basis for these thicker appearances is some experimental work; however, portionindications. “metallic”, The basis again for thesea departure thicker from appearances convention. is some Most experimental observers agree work; that however, there this work was never reviewed or published [25]. A second issue is the calling of the thicker maythis worknot be was a commonly-accepted never reviewed or published color described [25]. A as second “metallic”. issue isFurther, the calling this ofcolor the may thicker be portion “metallic”, again a departure from convention. Most observers agree that there confusedportion “metallic”, with silver again sheen. a departureSeveral parties from have convention. questioned Most this observers code [26,27]. agree The that current there may not be a commonly-accepted color described as “metallic”. Further, this color may be ASTMmay not standard be a commonly-accepted has the conventional color thin described appearance as “metallic”. codes. There Further, have been this color no scientific may be confused with silver sheen. Several parties have questioned this code [26,27]. The current papersconfused supporting with silver the sheen. thicker Several portions parties of the have Bonn questioned code. There this is code strong [26 evidence,27]. The that current the ASTM standard has the conventional thin appearance codes. There have been no scientific ASTMclearest standard signal for has thickness the conventional is the rainbow thin appearance sheen, as shown codes. in There Figure have 2. been no scientiﬁc papers supporting the thicker portions of the Bonn code. There is strong evidence that the papersAn supporting important thequestion thicker is portionsthe determination of the Bonn of code. the relevant There is thickness strong evidence ranges. that Figure the clearest signal for thickness is the rainbow sheen, as shown in Figure 2. 3clearest illustrates signal the for orders thickness of magnitude is the rainbow relevant sheen, to an as oil shown spill situation. in Figure2 An. important point An important question is the determination of the relevant thickness ranges. Figure is thatAn the important easily determined question is visual the determination ranges are orders of the relevantof magnitude thickness below ranges. the relevant Figure3 3 illustrates the orders of magnitude relevant to an oil spill situation. An important point thicknessillustrates ranges the orders needed of magnitudefor oil spill countermea relevant to ansure oil considerations. spill situation. This An importantfigure also pointillus- is that the easily determined visual ranges are orders of magnitude below the relevant tratesis that the the errors easily that determined would occur visual if one ranges used are incorrect orders data of magnitude on thickness below to determine the relevant the thickness ranges needed for oil spill countermeasure considerations. This figure also illus- volume.thickness The ranges errors needed could be for as oil large spill as countermeasure orders of magnitude. considerations. This ﬁgure also trates the errors that would occur if one used incorrect data on thickness to determine the illustratesTwo codes the errors are thatcurrently would used, occur that if one of used the incorrectASTM code data and on thicknessthe Bonn to Agreement determine volume. The errors could be as large as orders of magnitude. Codes.the volume. These The two errors codes could cover be widely as large different as orders thickness of magnitude. regimes, as shown in Figure 4. Two codes are currently used, that of the ASTM code and the Bonn Agreement Codes. These two codes cover widely different thickness regimes, as shown in Figure 4.

Figure 3. Thicknesses of interest to the spill responder.responder.

TwoFigure codes 3. Thicknesses are currently of interest used, to that the of spill the responder ASTM code. and the Bonn Agreement Codes. These two codes cover widely different thickness regimes, as shown in Figure4.

Figure 4. The comparison of the ASTM and Bonn Color Codes. Figure 4. The comparison of the ASTM and Bonn Color Codes. Figure 4. The comparison of the ASTM and Bonn Color Codes.

J. Mar. Sci. Eng. 2021, 9, 97 9 of 13

3. Visibility of Discharges Just as thin slicks on water appear to show consistency in appearance, discharges from vessels were thought to show consistency in appearance [2]. The proposition was that appearance of oil on the surface related to the oil content of discharge at a given speed. For example, it was proposed that discharges below 50 or 100 ppm could not be observed if the vessel was moving at 8 knots or more. It was thought that the greater the vessel speed, the greater the discharge could be before the slick was visible. These speed–discharge relationships actually depended both on the appearance of the resulting slick as well as the rise time and spreading of the oil. Because the speed of deposition was similar for most vessels, consistency in the observation of oil slicks was thought to be relevant. The purpose of this speculation and subsequent tests was to establish a maximum oil discharge value at sea, which did not result in a visible slick. These types of test are subject to high variability, much more than that for slick appearance, as listed in Section 2.1 above. Several tests were conducted and these tests ceased when discharge limits were established by a United Nations organization independently of these tests [28]. Early visibility test results are summarized in Table5.

Table 5. Minimum discharge visible with win and speed of discharge (Data from Fingas et al., 1999 [2]).

Wind Speed of the Discharge Vessel in Knots (and Visibility Minimum in ppm) Beaufort 2 4 6 8 12 16 0 50 - - 100 - 100 0 100 - 250 250 250 - 1 50 50 50 75 150 200 1 - - - 100 - - 2 67 87 100 200 - - 3 - 200 - 100 - 8500 4 - - 300 300 - 8500 5 4500 4500 4500 2300 - - 6 - 8000 - 4500 - -

4. Determination of Water-in-Oil Emulsification Water-in-oil emulsions sometimes form in rough waters after oil products are spilled. These emulsions, often called “chocolate mousse” or “mousse” by oil spill workers, can make the cleanup of oil spills very difficult [29]. When water-in-oil emulsions form, the physical properties of oil change dramatically. As an example, stable emulsions contain from 60 to 80% water, thus expanding the spilled material from 2 to 5 times the original volume. Most importantly, the viscosity of the oil typically changes from a few hundred mPa·s to about 100,000 mPa·s, an increase by a factor of 500 to 1000. A liquid product is changed into a heavy, semi-solid material. There are several types of emulsions, two of which are colored reddish because of the specular reflection of light from the water droplets in the oil. These colors can be seen from the air and are distinctive when most of the oil in the scene is emulsified in this manner [4]. This is shown in Figure5. It should be noted that water-in-oil emulsions sometimes form, dependent on various conditions [30]. It is important to note that the red color of water-in-oil emulsions is caused by the specular reflection of the emulsion. Water-in-oil emulsions are opaque to light. The thickness of water-in-oil emulsions is unknown but is thought to be much greater than sheen [30]. J. Mar. Sci. Eng. 2021, 9, x FOR PEER REVIEW 10 of 13

J. Mar. Sci. Eng. 2021, 9, 97 10 of 13 J. Mar. Sci. Eng. 2021, 9, x FOR PEER REVIEW 10 of 13

Figure 5. Emulsion shown from an aircraft from an altitude of approximately 100 m.

Fingas and Fieldhouse [30] found that four clearly-defined water-in-oil types are formed by crude oil when mixed with water. The types are named stable water-in-oil Figureemulsions,Figure 5. 5.Emulsion Emulsion meso-stable shownshown water-in-oil fromfrom an an aircraft aircraft emulsions, from from an an altitude altitudeentrained of of approximately approximately water, and unstable 100 100 m. m. water-in-oil types. The differences among the four types are quite large [31]. The overall properties of the water-in-oilFingasFingas andand types FieldhouseFieldhouse are shown [30[30]] foundinfound Table thatthat 6. Table fourfour clearly-deﬁned6clearly-defined shows that the water-in-oil water-in-oilbiggest difference typestypes arebe-are formedtweenformed types byby crudecrude are shown oiloil whenwhen in the mixedmixed typical with with lifetime water. wate ofr.The theThe water-in-oil typestypes areare named named types. stable Meso-stablestable water-in-oil water-in-oil emul- emulsions,sionsemulsions, appear meso-stable meso-stable very much water-in-oil water-in-oil like stable emulsions,emulsions emulsions, but entrained entrained break water,down water, within and and unstable unstable a few days. water-in-oil water-in-oil Figure types.6types. shows TheThe a typical differencesdifferences scenario among where the emulfour four sionstypes types appearar aree quite quite in largestreaks large [31]. [31 among]. The The overallsheen. overall properties properties of ofthe the water-in-oilThis water-in-oil shows typesthat types theare are reddish-brownshown shown in inTable Table characte6.6 Table. Table ristic66 shows shows of water-in-oil that that the the biggest biggest emulsions, difference difference meso- be- betweenstabletween andtypes types stable, are are shown can shown be in visually the in the typical typicalidentified. lifetime lifetime It ofshould the of water-in-oil the be water-in-oilnoted that types. the types. Meso-stablepresence Meso-stable of emul-emul- emulsionssionssions negatesappear appear veryother verymuch appearance much like stable like information; stable emulsions emulsions however, but break but sheen breakdown or downthinwithin slicks within a few can days. aexist few along-Figure days. Figureside6 shows emulsion,6 shows a typical aas typical shownscenario scenario in where Figure where emul 6. Emulsisions emulsions onappear does appear innot streaks form in streaks inamong thin among slicks sheen. or sheen. sheen. This shows that the reddish-brown characteristic of water-in-oil emulsions, meso- stable and stable, can be visually identified. It should be noted that the presence of emulsions negates other appearance information; however, sheen or thin slicks can exist along- side emulsion, as shown in Figure 6. Emulsion does not form in thin slicks or sheen.

FigureFigure 6.6. PhotographPhotograph showingshowing emulsionemulsion streaksstreaks surroundsurround by by sheen sheen (thin (thin oil). oil). This This photograph photograph was takenwas taken from from an aircraft an aircraft ﬂying flying at about at about 1000 m.1000 m.

Figure 6. Photograph showing emulsion streaks surround by sheen (thin oil). This photograph was taken from an aircraft flying at about 1000 m.

J. Mar. Sci. Eng. 2021, 9, 97 11 of 13

Table 6. Four ways by which oil uptakes water.

J. Mar. Sci. Eng. 2021, 9, x FOR PEER REVIEW Type Color Typical Lifetime11 of 13

Entrained as oil, typically black, shiny 2 to 7 days Meso-stable reddish until broken 2 to 7 days Table 6. Four waysStable by which oil uptakes water. reddish months Unstable or does not uptake water as oil, typically black - Type Color Typical Lifetime Entrained as oil, typically black, shiny 2 to 7 days This shows that the reddish-brown characteristic of water-in-oil emulsions, meso- Meso-stable reddish until broken 2 to 7 days stable and stable, can be visually identiﬁed. It should be noted that the presence of Stable reddish months emulsions negates other appearance information; however, sheen or thin slicks can exist alongsideUnstable emulsion, or does not as shown uptake in water Figure 6. Emulsion as oil, typically does not black form in thin slicks - or sheen.

5.5. ObservationObservation ofof Oil-in-WaterOil-in-Water Emulsiﬁcation Emulsification SpillSpill countermeasures countermeasures sometimes sometimes involve involve the the application application of of chemical chemical dispersants dispersants to to thethe oil. oil. The The results results of of such such activity activity may may result result in a in dispersion a dispersion of droplets of droplets under under the oil the slick. oil Thisslick. oil-in-water This oil-in-water emulsion emulsion is called is called a dispersion a dispersion in the in oil the spill oil trade. spill trade. If dispersion If dispersion is suc- is cessful,successful, a coffee-colored a coffee-colored plume plume in the in water the water is formed, is formed, such as such that as shown that shown in Figure in7 Figure[5,32]. 7 This[5,32]. is This a result is a result of the of light the light reﬂection reflection from from the manythe many 5 to 5 50to 50µm µm droplets droplets in in the the water water columncolumn [[33].33]. The coffee-colored pl plumeume is is used used as as a asign sign of of di dispersantspersant effectiveness effectiveness by by oil oilspill spill cleanup cleanup personnel. personnel. This This coffee-colored coffee-colored plume plume develops develops after after about about 20 20min min after after the theapplication application of dispersants, of dispersants, slowly slowly dissipates dissipates and and is generally is generally visible visible up to up about to about 3 h after 3 h aftera spill a spilldispersion. dispersion. The appearance The appearance of oil-in-water of oil-in-water emulsions emulsions is dependent is dependent on their on theirbeing beingno oil no on oil the on water’s the water’s surface surface to block to blockthe tran thesmission transmission of light of into light and into out and of out the of water the watercolumn. column.

FigureFigure 7. 7.The The appearance appearance of of a a coffee-colored coffee-colored plume plume after after the the application application of of a a chemical chemical dispersant dispersant to crudeto crude oil inoil a in test a test tank. tank. This This photo photo was was taken taken about about 20 min 20 min after after the applicationthe application of a dispersant.of a dispersant. This This coffee-colored plume is the result of light reflection from droplets of oil in the water in the coffee-colored plume is the result of light reﬂection from droplets of oil in the water in the size range size range of 5 to 50 µm. of 5 to 50 µm.

6.6. OtherOther ObservationsObservations SeveralSeveral other other observations observations of of oil oil slicks slicks may may have have utility utility to to the the spill spill countermeasures countermeasures efforteffort [6[6].]. ThisThis includesincludes obviousobvious physicalphysical parametersparameters suchsuch asas slickslick position,position, movement,movement, aerialaerial coveragecoverage andand distribution distribution or or patchiness patchiness [ 34[34].]. Additional Additional informationinformation thatthat may may be be sought is the viscosity of the oil, although there is no clear evidence of viscosity indications in any documented observations.

7. Conclusions The use of color and other visual observations has been used for many years to characterize oil on the sea. Visual methods to estimate oil thickness are restricted by physics

J. Mar. Sci. Eng. 2021, 9, 97 12 of 13

sought is the viscosity of the oil, although there is no clear evidence of viscosity indications in any documented observations.

7. Conclusions The use of color and other visual observations has been used for many years to characterize oil on the sea. Visual methods to estimate oil thickness are restricted by physics to thicknesses smaller than those of rainbow sheens (<3 µm). One can observe that some slicks are not sheen and are probably thicker. The use of color codes to estimate thickness should be restricted to that of rainbow sheens and lesser thicknesses (<3 µm). Other slick thicknesses do not have a basis in physics, nor in practical observations. Many different codes were developed, largely based on older work, and some with newer experimental work. Although there is good consistency for thin slicks, that is, for sheen and rainbow sheen, there are no consistencies for the appearance of thicker slicks (>3 µm). The reddish color of some water-in-oil emulsion types enables the detection of this oil condition. This is limited to stable or nearly-stable water-in-oil emulsions. The coffee- colored reﬂection of light from 5 to 50 µm oil droplets in water from a chemically-dispersed oil slick can give an indication of this dispersion. The discharge of oil from ships cannot be quantitatively monitored using the appearance of oil on the surface.

Funding: This research received no external funding. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Conﬂicts of Interest: The author declares no conﬂict of interest.

References 1. Fingas, M.F. The Challenges of Remotely Measuring Oil Slick Thickness. Remote Sens. 2018, 10, 319. [CrossRef] 2. Fingas, M.F.; Brown, C.E.; Gamble, L. The Visibility and Detectability of Oil Slicks and Oil Discharges on Water. In Proceedings of the Twenty-Second Arctic and Marine Oil Spill Program Technical Seminar, Calgary, AB, Canada, 2–4 June 1999; pp. 865–886. 3. Sicot, G.; Lennon, M.; Miegebielle, V.; Dubucq, D. Estimation of the thickness and emulsion rate of oil spilled at sea using hyperspectral remote sensing imagery in the SWIR domain. Int. Arch. Photogram. Remote Sens. Spat. Inf. Sci. ISPRS Arch. 2015, 40, 445–450. [CrossRef] 4. Lu, Y.; Shi, J.; Hu, C.; Zhang, M.; Sun, S.; Liu, Y. Optical interpretation of oil emulsions in the ocean—Part II: Applications to multi-band coarse-resolution imagery. Remote Sens. Environ. 2020, 242, 111778. [CrossRef] 5. IPECA. Dispersants: Surface Application; International Association of Oil & Gas Producers: London, UK, 2015. 6. De Padova, D.; Mossa, M.; Adamo, M.; De Carolis, G.; Pasquariello, G. Synergistic use of an oil drift model and remote sensing observations for oil spill monitoring. Environ. Sci. Pollut. Res. 2017, 24, 5530–5543. [CrossRef][PubMed] 7. US Congress. Report on Oil-Pollution Experiments—Behaviour of Fuel Oil on the Surface of the Sea; US Congress: Washington, DC, USA, 1930. 8. Cheemalapati, S.; Orth, H.P.F.; Wang, H.; Konnaiyan, K.R.; Morris, J.M.; Pyayt, A.L. Measurement of thickness of highly inhomogeneous crude oil slicks. Appl. Opt. 2017, 56, E72–E76. [CrossRef][PubMed] 9. Otremba, Z. Modeling of visibility of oil in the sea. Proc. SPIE 2002, 4881, 484–492. [CrossRef] 10. Otremba, Z.; Piskozub, J. The modification of light flux leaving a wind-roughened, oil covered sea surface example of computa- tions for shallow seas. Ocean. Stud. 2000, 29, 117–133. 11. Otremba, Z. A thin oil film covering the sea surface as a modifier of the downward transmission of light. Oceanologia 1997, 39, 397–411. 12. Pedrotti, F.L.; Pedrotti, L.S.; Pedrotti, L.M. Introduction to Optics, 3rd ed.; Prentice Hall: Upper Saddle River, NJ, USA, 2007. 13. Hornstein, B. The Appearance and Visibility of Thin Oil Films on Water; Environmental Agency Protection Report, EPA-R2-72-039; EPA: Cincinnati, OH, USA, 1972. 14. Otremba, Z. Oil droplets as light absorbents in seawater. Opt. Express 2007, 15, 8592–8597. [CrossRef][PubMed] 15. Horstein, B. The Visibility of Oil-Water Discharges. In Proceedings of the 1973 International Oil Spill Conference; American Petroleum Institute: Washington, DC, USA, 1973; pp. 91–99. 16. ASTM. Visually Estimating Oil Spill Thickness on Water; American Society for Testing and Materials: Conshohoken, PA, USA, 2016. 17. Allen, A.A.; Schlueter, R.S. Natural Oil Seepage at Coal Oil Point, Santa Barbara, California. Science 1970, 170, 974–977. [CrossRef] 18. API. Manual on Disposal of Refinery Wastes, Volume on Liquid Wastes; American Petroleum Institute: Washington, DC, USA, 1969. J. Mar. Sci. Eng. 2021, 9, 97 13 of 13

19. Parker, H.D.; Cormack, D. Evaluation of Infrared Line Scan (IRLS) and Side-Looking Airborne Radar (SLAR) over Controlled Oil Spills in the North Sea; Warren Spring Laboratory Report; Plenum Press: New York, NY, USA, 1979. 20. ITOPF (International Tanker Owners Pollution Federation). Aerial Observation of Oil at Sea; International Tanker Owners Pollution Federation: London, UK, 2005. 21. MacDonald, I.; Guinasso, N.L., Jr.; Ackleson, S.G.; Amos, J.F.; Duckworth, R.; Sassen, R.; Brooks, J.M. Natural Oil Slicks in the Gulf of Mexico Visible from Space. J. Geophys. Res. 1993, 98, 16351–16364. [CrossRef] 22. Brown, H.M.; Baschuk, J.J.; Goodman, R.H. The Limits of Visibility of Spilled Oil Sheens. In Proceedings of the Twenty-First Arctic and Marine Oil Spill Program (AMOP) Technical Seminar, Edmonton, AB, Canada, 10–12 June 1998; Environment Canada: Calgary, AB, Canada, 1998; Volume 2, pp. 805–810. 23. Canadian Coast Guard. Appearance and Thickness of an Oil Slick; Section 3, Annex C, Operations Manual; Canadian Coast Guard: Ottawa, ON, Canada, 1996. 24. Bonn Agreement. Guidelines for Oil Pollution Detection: Investigation and Post Flight Analysis/Evaluation for Volume Estimation. 2017. Available online: https://www.bonnagreement.org/ (accessed on 16 November 2019). 25. Lewis, A. The Use of Colour as a Guide to Oil Film Thickness: Phase I—A Literature Review; SINTEF Report No. STF66-F97075; SINTEF: Trondheim, Norway, 2000. 26. Lehr, W.J. Visual Observations and the Bonn Agreement. In Proceedings of the Thirty-Third Environment Canada Arctic and Marine Oil Spill Program Technical Seminar (AMOP), Halifax, NS, Canada, 7–9 June 2010; pp. 669–678. 27. Lehr, W.J.; Simecek-Beatty, D.; Leifer, I. Calculating Surface Oil Volume during the Deepwater Horizon Spill. In Proceedings of the Thirty-Fourth Environment Canada Arctic and Marine Oil Spill Program, Banff, AB, Canada, 4–6 October 2011; pp. 259–269. 28. Interpol. Marpol 73/78 Illegal Oil Discharges from Vessels; Interpol: London, UK, 2020. 29. NAS. Oil in the Sea III, Inputs, Fates and Effects, National Research Council; National Academies Press: Washington, DC, USA, 2002. 30. Fingas, M.; Fieldhouse, B. Studies on Crude Oil and Petroleum Product Emulsions: Water Resolution and Rheology. Colloids Surf. A 2009, 333, 67–81. [CrossRef] 31. Fingas, M. Models for Water-in-Oil Emulsion Formation. In Oil Spill Science and Technology; Elsevier: Amsterdam, The Netherlands, 2011; Chapter 10; pp. 243–273. 32. Cedre, F. (Ed.) Using Dispersants to Treat Oil Slicks at Sea: Airborne and Shipborne Treatment Response Manual; Centre for Documenta- tion, Research and Experimentation on Accidental Water Pollution (CEDRE): Brest, France, 2005. 33. Fingas, M. Oil Spill Dispersants: A Technical Summary. In Oil Spill Science and Technology; Gulf Publishing Company: Houston, TX, USA, 2011; Chapter 15; pp. 435–582. 34. Lee, M.-S.; Park, K.-A.; Lee, H.-R.; Park, J.-J.; Kang, C.-K.; Lee, M. Detection and dispersion of oil spills from satellite optical images in a coastal bay. Int. Geosci. Remote Sens. Symp. 2016, 7729643, 2491–2494. [CrossRef]